Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Generalized slowing in demented and cognitively-impaired-not-demented individuals Peters, Kevin Ross 1999

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1999-0588.pdf [ 3.8MB ]
Metadata
JSON: 831-1.0088987.json
JSON-LD: 831-1.0088987-ld.json
RDF/XML (Pretty): 831-1.0088987-rdf.xml
RDF/JSON: 831-1.0088987-rdf.json
Turtle: 831-1.0088987-turtle.txt
N-Triples: 831-1.0088987-rdf-ntriples.txt
Original Record: 831-1.0088987-source.json
Full Text
831-1.0088987-fulltext.txt
Citation
831-1.0088987.ris

Full Text

GENERALIZED SLOWING IN DEMENTED AND COGNITIVELY-rMPATRED-NOT-DEMENTED INDIVIDUALS by KEVIN ROSS PETERS B A . , Brock University, 1997 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES Department of Psychology We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1999 © Kevin Ross Peters, 1999 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT The purpose of the present investigation was to determine whether a widely accepted theory of normal cognitive aging can explain cognitive deficits in two groups of individuals with cognitive impairment. To answer this question, the Generalized Slowing Hypothesis (Birren, 1974; Cerrella, Poon, & Williams, 1980; Salthouse, 1980), which is that age-associated declines in high-level cognition are mediated by reductions in processing speed, was examined in 16 dementia (M age = 69; SD = 13.01) and 35 Cognitively-Impaired-Not-Demented (CIND) patients (M age = 64; SD = 9.65). Participants were recruited from the Clinic for Alzheimer's Disease and Related Disorders at the University of British Columbia Hospital. The California Verbal Learning Test (CVLT) and Rey-Osterrieth Complex Figure (ROCF) were used as measures of high-level cognition. Processing speed was measured by three tests, each of which has been demonstrated to represent unique components of processing speed: Finger Tapping, Simple Reaction, and Card Sorting (Graf & Uttl, 1995). Hierarchical regression analyses were performed in order to determine the ability of age to predict performance on the CVLT and ROCF before and after statistically controlling for the influence due to the three measures of processing speed. The results obtained in this investigation did not provide support for generalized slowing of processing in these two patient groups. The attenuation in the ability of age to predict performance after partialling out the influence due to processing speed was above 60% only for the performance of the CIND group on the CVLT, which is slightly lower than the lowest magnitude of attentuation previously reported in healthy adults. The attenuation in the predictive ability of age for performance on the ROCF was only 43% in the CIND group, and it was only 27% for the performance on the CVLT in the dementia group. These findings, although preliminary, suggest that the cognitive deficits of dementia and CIND patients are not merely the consequence of an acceleration of normal aging. Clearly, more research, with larger sample sizes, needs to be conducted to examine the tenability of generalized slowing in these two patient groups. iv TABLE OF CONTENTS ABSTRACT ii LIST OF TABLES vi LIST OF FIGURES vii ACKNOWLEDGMENTS viii CHAPTER ONE: Introduction and Overiew 1 CHAPTER TWO Generalized Slowing Hypothesis of Age-Associated Cognitive Decline 5 Conceptual and Operational Definitions of Processing Speed 7 Support for the Generalized Slowing Hypothesis 14 Generalized Slowing in Dementia and CIND Patients 15 Rationale 15 Processing Speed in Dementia and CIND Patients 17 Purpose and Hypotheses of the Present Investigation 18 CHAPTER THREE Method 21 Clinic Assessment 21 Participant Selection 24 Participants 25 Instruments 35 Procedure 40 V Data Preparation and Analysis 40 CHAPTER FOUR Results 45 Performance on Processing Speed and High-Level Cognitive Test Instruments .... 45 Hierarchical Regression Analyses 56 CHAPTER FIVE Discussion 66 Theoretical and Practical Implications Raised By This Research 68 Limitations of This Research 72 REFERENCES 75 APPENDIX A 83 APPENDIX B 87 vi LIST OF TABLES Table 1. Commonly used instruments to measure processing speed 8 Table 2. Demographic Information, Education, Depression, Mental Status, and Diagnostic Information of CIND and Dementia Participants 27 Table 3. Prescription Medication Usage by Category/Reason for CIND and Dementia Participants 31 Table 4. Vitamin, Mineral, Supplements, and Alternative Medicine Usage for CIND and Dementia Participants 33 Table 5. Means and Standard Deviations (in parentheses) for CIND and Dementia Participants on Processing Speed Tasks 46 Table 6. Mean Standardized Scores of CIND and Dementia Participants Performance Relative to Normative Data for Each Test Instrument 48 Table 7. Correlations and Coefficients of Determination (in parenthesis) Between Age and Performance of CVLT and ROCF for CIND and Dementia Participants 57 Table 8. Summary of Hierarchical Regression Analyses for Variables Predicting Performance on the California Verbal Learning Test in CIND Participants (N = 35) 61 Table 9. Summary of Hierarchical Regression Analyses for Variables Predicting Performance on the California Verbal Learning Test in Dementia Participants (N = 16) 62 Table 10. Summary of Hierarchical Regression Analyses for Variables Predicting Performance on the Rey-Osterrieth Complex Figure in CIND Participants (N = 35) 64 LIST OF FIGURES Figure 1. The mean performance of the CIND and dementia groups on the various CVLT measures 52 Figure 2. The mean performance of the CIND and dementia groups on the various ROCF measures 55 Figure 3. Scatterplot of the CVLT test scores as a function of age in the dementia group ... 59 Vll l ACKNOWLEDGMENTS First, I would like to express my gratitude to the patients that served as the participants in this investigation and their families. Indeed, this thesis would not have been possible without their willingness to cooperate in this investigation. Next, I would like to acknowledge the members of my Thesis Committee collectively and on an individual basis. I thank my Thesis Supervisor Peter Graf for allowing me the opportunity to work in his lab as one of his students. He has taught me a great deal over the past year and has always made sure that my writing made sense and contained the ideas that I wished to express. I am grateful to Sherri Hayden for providing me with the opportunity to work in the Alzheimer's Clinic and for teaching me about the important issues in clinical neuropsychology. I thank John Pinel for giving me advice on how to improve my public speaking abilities and for teaching me that "pauses are good." In addition, I would like to acknowledge the assistance of Bob Uttl, who developed the computerized processing speed measures that were used in this investigation, for providing me with his helpful advice in the administration of these measures and for his general knowledge of SPSS. My gratitude also goes out to the rest of the Graf Lab for providing me with useful feedback on earlier proposals of this research. I would also like to thank the staff at the Clinic for Alzheimer's Disease and Related Disorders for helping me over the past year. Last, but most certainly not least, I would like to thank my wife, Anne-Marie, for her support, patience, and understanding over the past three years that we have known each other. She was always willing to do more than her fair share of work when I had deadlines to meet or was just swamped. She also knew when to kindly let me know that I was losing perspective on things. For this I am very grateful. I hope that someday I am able to do the same for her in her academic future. Generalized Slowing in 1 CHAPTER ONE Introduction and Overview What is the nature of the relationship between normal aging and dementia? Are the processes of normal age-associated cognitive decline also present in individuals with cognitive impairment? How can we identify, at an early stage, which individuals in the population are going to develop dementia? These were the questions that provided the impetus for this investigation. Although the answers to these questions are currently unknown, they are becoming increasingly important as the proportion of elderly individuals in our society continues to grow. The purpose of the present research was to determine whether a widely accepted theory of normal cognitive aging can account for the cognitive deficits in two groups of patients with cognitive impairment. The Generalized Slowing Hypothesis (Birren, 1974; Cerrella, Poon, & Williams, 1980; Salthouse, 1980, 1985, 1996a) is the theory of normal cognitive aging that was the basis of this investigation. This hypothesis is that as we get older, the speed at which our central nervous system operates begins to slow down. One of the consequences of this slowing is that our ability to process information also decreases, and it is this decrease that produces the cognitive decline that is observed in the normal elderly. Support for this hypothesis comes from studies in which the ability of age to predict performance on tests of high-level cognition are reduced by up to 90% after statistically controlling for measures of processing speed. The two patient groups to which this hypothesis was applied in the present investigation were (a) dementia, and (b) Cognitively-Impaired-Not-Demented (CIND) individuals. Diagnostically, the difference between these two patient groups is the degree of Generalized Slowing in 2 cognitive impairment. A minimal criterion for the diagnosis of dementia is the presence of significant impairment in two or more cognitive domains (e.g., memory plus language, attention, visuospatial abilities, reasoning, or problem-solving). In contrast, the diagnosis of CIND is reserved for individuals demonstrating either (a) significant impairment in only one cognitive domain, or (b) milder impairment in one or more cognitive domains with the stipulation that the degree of impairment is not severe enough to warrant a diagnosis of dementia. Studies have shown that a higher percentage of CIND individuals than healthy adults will develop dementia (Bowen, Terri, Kukull, McCormick, McCurry, & Larson, 1997; Graham, Rockwood, Beattie, Eastwood, Gauthier, Tuokko, & McDowell, 1997; Tierney, Szalai, & Snow, 1996). Thus, CIND may represent a transitional or prodromal phase of dementia. Many studies are currently being conducted to find ways of identifying which CIND individuals will progress to dementia (Devand, Folz, Gorlyn, Moeller, & Stern, 1997; Flicker, Ferris, & Reisberg, 1991, 1993; Jacobs, Sano, Dooneief, Marder, Bell, & Stern, 1995; Masur, Sliwinski, Lipton, Blau, & Crystal, 1994). The focus of many of these studies is on identifying which neuropsychological tests best predict the future development of dementia in CIND individuals. To date, this approach has met with little success; no single neuropsychological test has been identified which consistently predicts this outcome. I believe that a better strategy is to focus instead on the variables that predict impaired performance on the neuropsychological tests themselves, as these variables may provide insight into the processes that lead to cognitive impairment. For example, many studies have shown that age is a significant predictor of performance on tests of high-level cognition (see Salthouse, 1985, 1991a, 1996a for reviews), and that-consistent with the Generalized Generalized Slowing in 3 Slowing Hypothesis—performance on measures of processing speed mediate the relationship between age and performance on tests of high-level cognition (see Salthouse, 1996a for review). Applying these findings to the present context, it has been known for quite some time that age is a risk factor for the development of dementia (Lezak, 1995). However, there have been no published reports of the degree to which deficits in processing speed mediate the relationship between age and performance on high-level cognitive tasks in demented or CIND patients. The present manuscript is an interim report of a study that is still being conducted. The sample sizes that have been obtained thus far are insufficient to provide enough power to adequately examine the Generalized Slowing Hypothesis in these two patient groups. The results reported in this investigation, and the inferences drawn from them, are to be treated as preliminary and exploratory. More reliable conclusions await larger sample sizes. This remainder of this manuscript is divided into four chapters. The purpose of the next chapter—Chapter Two—is to provide a summary of the Generalized Slowing Hypothesis and the rationale for examining this hypothesis in dementia and CIND patients. It is broken down into two sections. The first section begins by describing the Generalized Slowing Hypothesis and why this hypothesis has become so widely accepted in the cognitive aging literature. The manner in which processing speed—the central construct in the Generalized Slowing Hypothesis—has been operationally defined is then presented. This is followed by an overview of the main sources of support for the Generalized Slowing Hypothesis. The second section begins by providing a rationale for assessing this hypothesis in demented and CIND patients. A summary of the literature relevant to processing speed in these two patient groups is then presented. This section ends by briefly describing the manner in which the Generalized Slowing in 4 present investigation assessed the Generalized Slowing Hypothesis. Chapter Three describes how the present study was conducted. More specifically, the chapter contains information regarding the following: the clinical assessment procedures, diagnostic criteria, how the participants were selected, a description of the participants, and the instruments and research procedures that were employed in the present investigation. The results of the present research are reported in Chapter Four. In Chapter Five, the results of this investigation are interpreted with respect to the hypotheses that were described in Chapter Two. In addition, the theoretical and clinical implications that are raised by this investigation are discussed. Generalized Slowing in CHAPTER TWO Generalized Slowing Hypothesis of Age-Associated Cognitive Decline The Generalized Slowing Hypothesis is a widely accepted view of cognitive aging (Birren, 1974; Cerrella, Poon, & Williams, 1980; Salthouse, 1980, 1985, 1996a). It was proposed to explain age-associated decreases in those aspects of cognition that have been referred to as type A (Hebb, 1942), fluid (Horn & Cattell, 1963), or mechanic (Baltes, Dittmann-Kohli, & Dixon, 1984), and include measures of reasoning, problem-solving, memory, and visuospatial abilities (see Salthouse, 1985, 1991a; Verhaeghen & Salthouse, 1997 for reviews). This hypothesis is that "the central nervous system is functioning at a slower rate in older adults, [and consequently] mental operation time may be the principle mechanism behind age differences in nearly all aspects of cognitive functioning" (Salthouse, 1980, p. 61). There are two reasons why the Generalized Slowing Hypothesis has become widely accepted. The first is that it represents a biologically plausible explanation of normal age-associated cognitive decline. For example, age-associated reductions in synaptic density (Adams, 1987; Cragg, 1975; Huttenlocher, 1979; Masliah, Mallory, Hansen, DeTeresa, & Terry, 1993), demyelination of axons (Miller, Alston, & Corsellis, 1980), alterations in the functioning of neurotransmitter systems (see DeKosky & Palmer, 1994 for a detailed review) and changes in molecular and receptor functioning (Magnusson & Cotman, 1993) could all produce decreases in the speed at which the central nervous system is able to function. The second reason is that a great deal of support has been provided for this hypothesis (Salthouse 1996a). A summary of the evidence in support of generalized slowing is described later in this section. Generalized Slowing in 6 How might generalized slowing of the central nervous system lead to decreased performance on high-level cognitive tasks? Generalized slowing has been defined in two ways. The first of these pertains to the capacity of the central nervous system to hold information in memory while simultaneously performing manipulations on that information, which is commonly referred to as working memory capacity (Baddeley, 1986; Baddeley & Hitch, 1973; Craik & Jennings, 1992). The second pertains to the speed at which the central nervous system can process information, which is commonly referred to as processing speed (Birren, 1974; Cerrella et al., 1980; Salthouse, 1980, 1985, 1996a). Although both the working memory capacity and processing speed constructs have been pursued within the cognitive aging literature, the latter has recently received a greater amount of attention. One of the reasons for this increased focus on processing speed is that operational definitions of speed are less ambiguous than working memory capacity (Johnson & Rybash, 1993; Salthouse, 1985, 1991a). In addition, there is evidence that processing speed is a more basic and elementary construct that actually mediates the relationship between age and working memory capacity (Park, Lautenschlager, Smith, Earles, Frieske, Zwahr, & Gaines, 1996; Salthouse, 1991a,b, 1992a; Salthouse & Babcock, 1991). To illustrate this point, consider a task that requires several different pieces of information to be active simultaneously. In addition, assume that the speed at which each of those pieces is able to become active decreases with age. The end result would be that it takes longer to activate two or more pieces of information simultaneously, and therefore, it would take longer to perform the task (Salthouse, 1996a). The remainder of this chapter focuses on the processing speed construct. Generalized Slowing in 7 Conceptual and Operational Definitions of Processing Speed Processing speed is typically defined as the speed at which basic cognitive operations can be performed (Salthouse, 1985, 1996a). Many of the instruments that have been used to measure processing speed require participants to complete as many operations as possible within a specified period of time. Salthouse (1996a) has described seven tests that have been used frequently to measure processing speed in research on cognitive aging. Table 1 provides descriptive information for each of these tests. As can be seen in Table 1, the test-retest reliabilities of these test instruments are respectable (e.g., greater than .70), with the exception of Letter Comparison and Digit-Digit. Reliability is desirable because it allows us to be more confident that we are measuring the same construct on each occasion. Furthermore, the degree to which an instrument is reliable determines the degree to which that instrument can be correlated with other variables (Cohen & Cohen, 1983). In other words, the more reliable an instrument is, the more systematic variance that instrument has to potentially share with other variables or instruments. Given the variety of reliable test instruments described in Table 1, how does one go about choosing one of these instruments? To answer this question, one needs to examine the validity of these instruments. A test instrument is valid to the extent that it measures what it is intended to measure. In this case, a valid processing speed instrument would be one that actually measures processing speed. One of the ways to determine whether a test is valid is to identify, at a conceptual level, which component processes are required by that test. For example, at a minimum, most cognitive tests require the participant to perceive a stimulus (input component), cognitively process some aspect of that stimulus (cognition component), oo S3 o 13 (U N 13 >-. <u d O 0 0 1=1 13 CD CD O H co 0 0 .£ 'co 1/1 CD O o O H <u o oo • i-H co O N oo o <u <D a co O a o =3 O t - l o C o g i3 S- i O O o 00 a I (-1 1 3 CD CD O H • CO cd to CD O I "5b co O N o I o o o B CD a o H—» o o co ^3 00 O N 00 O co a g x CO CD CO C O O H co CD O CD O o 13 O "C CD O H 1 3 0 O o CD co CD X ! 13 CD •a 0> co CD bH O H CO co CD CT1 co O co O t-. o 1—1 0 0 13 •a •a E o o H-» CD CD X i co H-> co CD g =3 co X ! o a CD o s-l o 0 0 o t= o X> o C/l CD 13 CD o CD O H 'o CD 1 3 0 0 1=1 CD X i cd l-l 13 13 CD 'S CD CD xl H 0 0 .g CO 13 O •c CD O H 13 1=1 O o CD co O CO cd c 2 co CD co 1=1 O O H co CD o CD o o ( 3 co CD X o CO CO cd H-» 13 CD N •a CD E O H O O o 0 0 I H-» • >—I 0 0 Q o x> co i H-» 'oo 13 CD 'oo o CO cd is ' E H -f-» CD o "a CO D S I o o CD •B o O H O -*-> CD X j '3 CD X i H . £ "a, CO •3 CD •3 O H O o CD X i H-» o JD 13 13 CD 4=1 CD •B o T3 13 CD X j cd O H O x CO 0 0 CD t - l CD X H-» CD X I Cd O a 1 3 CD '3 CT1 CD t-C cd O H , i CO O - O O 1 •=H op -3 CD H—» <H-I o CD C3 O co CD J=l O 13 o CD O O co =3 ? H O o CD J=I CO cd CD J=l H CO =3 E O o CD 4 3 -t-» O CD X I CD 1 H-» CD X I "fl +-> 13 CD 13 o o Generalized Slowing in 12 and then respond to that stimulus (output component). This method of decomposing the component processes required by a given test instrument is referred to as a conceptual task analysis, and it allows one to assess the validity of a test instrument by identifying which processes are required for that task. In addition to providing a description of commonly used processing speed instruments and their test-retest reliabilities, Table 1 also provides the results of a conceptual task analysis of the component processes required for each of these instruments. As can be seen in Table 1, these instruments measure more than just processing speed. For example, they all involve some degree of vigilance, working memory, visuomotor coordination, and motor speed in addition to processing speed. Some of the instruments also involve visual scanning (Digit Symbol Substitution Test, Letter and Pattern Comparison, Digit-Symbol, and Digit-Digit) and comparison and decision making processes (Letter and Pattern Comparison, Digit-Symbol, and Digit-Digit). A consistent finding of this task analysis is that the paper-and-pencil tasks all involve one additional component process in the response that is required, (i.e., writing or drawing in addition to motor speed). This latter finding is important because it has been suggested that an ideal measure of processing speed is one in which the involvement of sensory (input) and motor (output) processes are minimized as much as possible (Graf & Uttl, 1995; Salthouse, 1991a, 1996a). The utility of this requirement becomes apparent when interpreting differences between groups on a given task. For example, if two groups differ in their performance oh a measure of processing speed, it would be desirable to attribute those differences to the speed of information processing rather than to the input or output processes required by the task. Therefore, it can be argued that the computerized tasks are preferable to paper-and-pencil Generalized Slowing in 13 tasks described here because the processes involved in the execution of the response are minimized. It has also been suggested that the cognitive processes involved in measures of processing speed should be as basic as possible (Salthouse, 1996a). Graf and Uttl (1995) conducted an interesting study in which they decomposed processing speed into the more basic and elementary components of motor speed, processing rate, and processing capacity. Their work was based on the analogy of the computer and the factors that affect its overall speed. A computer's speed is affected by at least two factors: its processing rate and bandwidth. Processing rate refers to the number of operations that the computer can perform per unit of time, which is typically expressed in MHz. Bandwidth or processing capacity refers to the number of items that a computer can operate on at any given time, which is typically expressed in bits. In their investigation, they operationally defined processing rate using a computerized simple reaction time task. As a measure of processing capacity, they used a computerized card sorting task in which there were either zero, four, or eight distractor items on each card. In addition to these two components of processing speed, they also measured the contribution of motor speed through the use of a finger tapping task. Their results indicated that these three components of processing speed made unique and significant contributions to explaining age-related variance in an episodic memory task. On the basis of these results, Graf and Uttl concluded that processing speed may be better conceptualized as system productivity or the amount of work that a system can perform per unit of time. Generalized Slowing in 14 Support for the Generalized Slowing Hypothesis A strong and reliable relationship has been shown to exist between measures of processing speed and performance on tests of high-level cognition across the life-span (for reviews see Salthouse, 1985, 1991a, 1996a). Perhaps, the strongest evidence for this relationship comes from studies in which hierarchical regression analyses and path analyses have been used to demonstrate the role of measures of processing speed in mediating the age-associated declines in high-level cognition. Hierarchical regression analyses have been useful in determining the amount of variance in high-level cognitive task performance accounted for by age before and after statistical control of measures of processing speed. The logic of these analyses is that age, by itself, is usually a very good predictor of test performance on high-level cognitive tasks. However, if the variance due to processing speed is partialled out, the ability of age to predict test performance drops considerably. The attentuation in the ability of age to explain test performance on high-level cognitive tasks after statistically controlling for the effect of processing speed provides evidence for a mediational role of processing speed in the age-cognition relationship. Salthouse (1992b) has provided a useful guide for interpreting the magnitude of attentuation that occurs in the ability of one independent variable to explain the variance in a dependent variable after the statistical control of a second independent variable. He classifies attenuations less than 20% as small, those between 20 to 40% as interesting, those between 40 and 60% as important, and those greater than 60% as major. In a review of 29 studies examining the relationship between measures of processing speed and performance on a variety of memory tasks, Salthouse (1996a) found that, when considered alone, age explained Generalized Slowing in 15 an average of 20% of the variance in the memory tasks. After statistically controlling for the effect due to measures of processing speed, the amount of variance explained by age dropped considerably, to an average of 2.8%. Thus, the average amount of attenuation in the effect of age was 86% ([20-2.8]/20 x 100 = 86). According to the guidelines suggested by Salthouse (1992b), this magnitude of attentuation is major, deserving of further consideration. Path analysis is a graphical means of presenting the results obtained from hierarchical regression analyses regarding the mediational role of certain variables. These models allow for the direction and magnitude of both direct and indirect relationships among various variables to be determined, such that only those relationships that are deemed to be significant remain in the model. Several such analyses have been conducted thus far and all have consistently demonstrated that processing speed mediates the relationship between age and various measures of high-level cognition (Bors & Forrin, 1995; Graf & Uttl, 1995; Lindberger, Mayr, & Kliegl, 1993; Park et al., 1996; Sliwinski & Buschke, 1997; Salthouse, 1991b, 1993, 1994a,b; Verhaeghen & Salthouse, 1997). Generalized Slowing in Dementia and CIND Patients Rationale Can the Generalized Slowing Hypothesis of normal cognitive aging be used to explain the cognitive impairments that are associated with clinical populations involving demented and CIND individuals? This question is interesting for both theoretical and clinical reasons. From a theoretical point of view, there is much debate in the literature as to whether normal and pathological cognitive aging (e.g., dementia) represent two distinct entities or whether they fall along a continuum such that the latter is merely an exaggeration or accelerated version of the former (Ferris & Kluger, 1996; Huppurt, Brayne, & O'Connor, Generalized Slowing in 16 1994; Rediess & Caine, 1996; & Smith, Petersen, Parisi, Ivnik, Kokmen, Tangalos, & Waring, 1996). Support for the Generalized Slowing Hypothesis in these two patient groups would provide evidence in favor of a continuum, suggesting that the same processes are involved in normal and pathological cognitive aging. Generalized slowing would be indicated by the finding of attenuations in the ability of age to explain performance on high-level cognitive tests after partialling out the influence due to processing speed that are equal to, or greater than, those observed in the normal elderly. However, failure to support the hypothesis (i.e., attenuations in the ability of age to explain performance on high-level cognitive tests after partialling out the influence due to processing speed that are less than, those observed in the normal elderly) would provide evidence against a continuum, suggesting that normal and pathological cognitive aging are two distinct entities. . From a clinical point of view, the relationship between processing speed and performance on tests of high-level cognition may serve as a useful means for the early identification of individuals who are in the process of developing a dementia. The CIND group is unique in that there is evidence demonstrating that the percentage of these individuals that will go on to develop a dementia is higher than in the normal population (12 to 24% compared to 8% of individuals over 65) (Bowen, Terri, Kukull, McCormick, McCurry, & Larson, 1997; Graham, Rockwood, Beattie, Eastwood, Gaufhier, Tuokko, & McDowell, 1997; Tierney, Szalai, & Snow, 1996). The relationship between processing speed and high-level cognition may be different in those CIND individuals who later develop dementia than in those who do not. Generalized Slowing in 17 A recent study conducted by Collins and Long (1996) found that measures of simple and choice reaction time were more sensitive than the Impairment Index of the Halstead Reitan Neuropsychological Test Battery in detecting mild cognitive impairment present in individuals following traumatic brain injury (TBI). They first determined cut-off scores for both reaction time measures using a mixed group of TBI patients and normal controls that resulted in correct classification rates of at least 94%. They then examined the classification rate of these reaction time cut-off scores in another group of TBI patients who, according to the traditional cut-off score of the Halstead Reitan Impairment Index, were no longer impaired. They found that over 60% of the patients who were classified as "recovered" according to the Impairment Index cut-off score were still impaired using the reaction time cut-off scores. Applying this finding to the present investigation, it is possible that measures of processing speed are more sensitive at detecting the onset of dementia than more complex neuropsychological measures. Earlier diagnosis would permit the implementation of earlier interventions, perhaps before irreversible brain damage has occurred. Processing Speed in Dementia and CIND Patients Although examination of the Generalized Slowing Hypothesis in dementia and CIND patients would be interesting from both theoretical and clinical points of view, to my knowledge there have been no published reports to date in which this hypothesis has been examined in either of these two groups. However, many studies have examined processing speed in dementia patients (Ferris, Crook, Sathananthan, & Gershon, 1976; Muller, Richter, Weisbrod, & Klingberg, 1991; Pate & Margolin, 1994; Nebes & Madden, 1988; Nebes & Brady, 1992; Nebes, Brady, & Reynolds, 1992; Pate, Margolin, Friedrich, & Bentley, 1994; Pirozzolo, Christensen, Ogle, Hansen, & Thompson, 1981; Sliwinski & Buschke, 1997; Generalized Slowing in 18 Williams, Jones, Briscoe, Thomas, & Cronin, 1991). A common finding in all of these studies has been that processing speed is significantly slower in dementia patients than in age-matched controls. However, most of these studies involved small sample sizes (e.g., fewer than 20 dementia patients), and more importantly, processing speed was treated as the dependent variable in all of these studies. In other words, the focus of these studies was on whether or not processing speed was slowed in dementia patients relative to age-matched controls, not whether processing speed could mediate the relationship between age and high-level cognition. Regarding the status of processing speed in CIND patients, there have been two studies demonstrating that performance on the WAIS-R Digit Symbol Substitution subtest is significantly lower in mildly impaired patients than in healthy controls. In addition, using a longitudinal design, Devand, Folz, Gorlyn, Moeller and Stern (1997) found that significant differences in performance on the WAIS-R Digit Symbol Substitution subtest were present at baseline between those patients who later developed dementia and those who did not. However, in the longitudinal study conducted by Flicker, Ferris, and Reisberg (1993), performance on this same measure was not significantly different at baseline between those patients who later developed dementia and those who did not. Clearly, more research needs to be conducted involving different measures of processing speed in CPND patients.. Purpose and Hypotheses of the Present Investigation The purpose of the present investigation was to determine whether the Generalized Slowing Hypothesis of normal cognitive aging could explain the cognitive deficits in demented and CIND individuals. To test this hypothesis, three measures of processing speed along with several measures of high-level cognition were administered to 25 dementia Generalized Slowing in 19 patients and 39 CIND patients from the Clinic for Alzheimer Disease and Related Disorders, located in the University of British Columbia Hospital. Three measures of processing speed were used: finger tapping, simple reaction time, and card sorting. The measures of high-level cognition consisted of the California Verbal Learning Test or CVLT (Delis, Kramer, Kaplan, & Ober, 1987) and the Rey-Osterrieth Complex Figure or ROCF (Rey, 1941; Osterrieth, 1944 both translated by Corwin & Bylsma, 1993). Hierarchical regression analyses were performed to assess the ability of the three processing speed measures to mediate the age-cognition relationship. In this manner it was possible to determine the amount of variance in CVLT and ROCF performance that was accounted for by age before and after statistically controlling for the influence due to measures of processing speed, thus providing a direct test of the Generalized Slowing Hypothesis. This method of analysis also allowed for the statistical control of other important variables such as sex, education, medication use, and depression. According to the Generalized Slowing Hypothesis, age-associated decreases in processing speed are responsible for the decline in high-level cognition that is observed in the normal elderly. There are two predictions that logically follow if generalized slowing is present in these two groups. First, after partialling out the influence due to processing speed, the attenuation in the predictive ability of age to explain performance on the CVLT and ROCF will be at least equal to the lowest magnitude of attentuation that has been observed in healthy adults. The critical magnitude of attenuation that was chosen for the present investigation was 60%. This value was chosen in accordance with the classification made by Salthouse (1992a) that reductions of 60% or greater represent "major" attenuations. In addition, in the review conducted by Salthouse (1996a), the lowest magnitude of attenuation Generalized Slowing in 20 in the predictive ability of age in explaining memory performance after statistically controlling for measures of processing speed was 63%. The second prediction was that the magnitude of attenuation in the predictive ability of age will be the same for both the CVLT and the ROCF instruments. The rational for this prediction was that if the slowing in these two patient groups is indeed generalized, then all domains of cognition should be affected equally. Generalized Slowing in 21 CHAPTER THREE Methods Clinic Assessment Participants were recruited from the Clinic for Alzheimer's Disease and Related Disorders, located in the University of British Columbia Hospital. This is a multidisciplinary clinic that provides assessment, diagnostic, and educational and counselling services to patients and their families. In addition, the clinic is actively involved in research projects concerning Alzheimer's disease and related neurodegenerative disorders. On the basis of subjective complaints of cognitive impairment, patients are referred to this clinic by their family physician for an initial assessment, which takes place over a period of several days. As part of the referral process, patients are given a Mini Mental State Exam (Folstein, Folstein, & McHugh, 1975) by either their family physician or by one of the clinic team members during their initial assessment. Each patient is required to fill out two informed consent forms at the beginning of their assessment. Signing these forms gives the clinic team members permission to obtain the patient's relevant medical and personal information and to use any of the information obtained during their assessment for future research purposes. A copy of these consent forms are provided in Appendix A. During their assessment, all patients undergo a comprehensive laboratory screen including routine blood work, neuroimaging with CT scan, electro-cardiogram, and chest x-ray. Each patient may be evaluated by clinic team specialists in the following areas: Geriatric Medicine, Medical Genetics, Neurology, Neuropsychology, and Social Work. Additional evaluation by the clinic Psychiatrist as well as Speech and Language Pathologist is conducted in those cases in which such evaluation is deemed Generalized Slowing in 22 necessary on the basis of the patient's presenting complaints and history. Each clinic team member gathers her or his own information through the use of formal testing procedures and/or a structured interview format with the patient and, whenever possible, a collateral informant: spouse, son, daughter or friend. After the patient has been evaluated by all relevant clinic team members, a meeting is held during which their case is reviewed by the clinic team and a consensus diagnosis is reached. The diagnostic process consists of two stages: (a) determining the presence or absence of dementia, and (b) further classification within the dementia and not demented categories. The diagnosis of dementia is made in accordance with the criteria set out in the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSMIV) (American Psychiatric Association, 1994). In addition to using the standard DSM IV criteria for dementia, the diagnostic process is aided through the use of the Functional Rating Scale (Tuokko & Crockett, 1989). The Functional Rating Scale (FRS) is a derivative of the multidimensional Clinical Dementia Rating Scale (Hughes, Berg, & Danzinger, 1982; for a discussion of the similarities between the FRS and other commonly used dementia criteria and rating scales see Tuokko & Crockett, 1991). The FRS has eight dimensions: Memory, Social/Occupational, Home and Hobbies, Problem Solving and Reasoning, Personal Care, Language Skills, Affect, and Orientation. For each of the eight FRS dimensions, an impairment rating of either healthy (1), questionable (2), mild (3), moderate (4) or severe (5) is assigned. The individual scores of these eight dimensions are summed together to obtain a total score. A total score between 16 and 24 represents a mild degree of impairment, scores between 25 and 32 represent moderate impairment, and scores greater than 32 represent severe impairment. Previous research in which the clinical staging of Alzheimer's disease Generalized Slowing in 23 was examined longitudinally has demonstrated that these impairment indices are clinically useful (Feldman, Schulzer, Wang, Tuokko, & Beattie, 1995). The criteria contained in the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer's Disease and Related Disorders Association (McKhann, Drachman, Folstein, Katzman, Price, & Stadlan, 1984) is used to further classify patients diagnosed with dementia into one of the following three categories: possible Alzheimer's disease, probable Alzheimer's disease, or dementia unlikely due to Alzheimer's disease (e.g., vascular dementia, frontotemporal dementia, and dementia with Lewy Bodies). Patients who do not meet criteria for dementia are further classified as either Not-Cognitively-Impaired (NCI) or CIND. A diagnosis of NCI is reserved for those cases in which there is no subjective or objective evidence of cognitive impairment, relative to age, evident within the assessment. Subjective evidence of cognitive impairment requires the reporting of such impairment by either the patient or an informed collateral. In contrast, objective evidence of cognitive impairment requires the presence of test performance that is interpreted by the clinic team to be at least one standard deviation below the patient's estimated premorbid level of functioning. The diagnosis of CIND requires either (a) significant impairment (i.e., performance that is two or more standard deviations below premorbid functioning) in only one cognitive domain, or (b) milder impairment (i.e., one standard deviation below premorbid functioning) in one or more cognitive domains with the stipulation that the degree of impairment in these domains not be sufficient enough to warrant a diagnosis of dementia. After the consensus diagnosis is obtained, the severity level of overall impairment for each patient is determined by assigning a score from the Global Deterioration Scale Generalized Slowing in 24 (Reisberg, Ferris, De Leon, & Crook, 1982). The Global Deterioration Scale (GDS) is a rating scale consisting of seven stages, ranging from Stage 1 (no cognitive decline) to Stage 7 (very severe cognitive decline). Those patients diagnosed as NCI receive a GDS score of 1. CIND patients receive scores of either 2 (very mild cognitive decline) or 3 (mild cognitive decline); whereas, dementia patients receive scores greater than 3. The CIND diagnosis in the present study encompasses conditions that other investigators have referred to as age-associated memory impairment or age-related cognitive decline (Albert, 1993; Blackford & LaRue, 1989; Crook, Bartus, Ferris, Whitehouse, Cohen, & Gershon, 1986; Howieson, Holm, Kaye, Oken, & Howieson, 1993; Ivnik, Malec, Smith, Tangalos, Petersen, Kokmen, & Kurland, 1992), mild cognitive impairment (Blackford & LaRue, 1989; Flicker, Ferris, & Reisberg, 1991; Larrabee, Levin, & High, 1986; Levy, 1994; Smith, Petersen, Parisi, Ivnik, Kokmen, Tangalos, & Waring, 1996), and mild neurocognitive disorder (Caine, Grossman, & Lyness, 1995; Morris, McKeel, Storandt, Rubin, Price, Grant, Ball, & Berg, 1991). Participant Selection The sample in the present investigation was one of convenience, rather than consisting of randomly selected or consecutively admitted patients. Only patients who had scores that were higher than 15 on the initial Mini Mental State Exam were approached for recruitment into the study. The principle investigator conducted the majority (67%) of the neuropsychological assessments while the remaining 33% of the assessments were performed by the psychometrist employed at the clinic. For those assessments conducted by the clinic psychometrist, the principle investigator attended, near the end of the session, to administer those additional tests that were part of the research test battery. Generalized Slowing in 25 Approximately half (51.6%) of the participants recruited for the present study were also involved in a Collaborative Cohort of Related Dementias (ACCORD) study that is being conducted at eight different sites across Canada (Feldman, Beattie, Hayden, Sadovnik, Studney, & Beach, 1996). The purpose of the ACCORD study is to examine the course and outcome of various conditions associated with cognitive impairment using a longitudinal design. The duration of the ACCORD study is 4.5 years, with participants reassessed annually. During the time of the present study, the participants obtained from the ACCORD study were being seen for their first reassessment. Only those ACCORD study participants who were recruited from the Vancouver site were considered for the present study. A series of independent sample t-tests were conducted to compare the portion of the sample that was composed of participants from the ACCORD study with those participants that were not from the ACCORD study. There were no significant differences between these two groups in terms of age, number of years of education, total score on the Geriatric Depression Scale (Brink, Yesavage, Lum, Heersema, Adey, & Rose, 1982; Yesavage, Brink, Rose, & Adey, 1986), age at symptom onset, age at first diagnosis and score on the Mini Mental State Exam (Folstein, Folstein, & McHugh, 1975). Therefore, the data obtained from the ACCORD and non ACCORD study participants were combined for all data analyses. Participants A total of 64 participants were recruited for the present study, with 25 (39.1%) receiving the diagnosis of dementia and 39 (60.9%) being diagnosed as not demented. The breakdown of the dementia patients were as follows: 12 with possible Alzheimer's disease (18.8% of total sample), eight with probable Alzheimer's disease (12.5% of total sample) and Generalized Slowing in 26 five with unlikely Alzheimer's disease (7.8% of total sample). Al l of the patients in the not demented category were diagnosed as CIND. Of the 64 participants, 13 (nine dementia and four CIND) were excluded from farther analyses due to incomplete data. The full set of tests were not given to these participants, either at the request of the participant or at the discretion of the psychometrist. For example, eight of the participants wished to discontinue testing, and there were five instances where, prior to administering the test, the psychometrist didn't think the participant would fully understand the task. This left a total of 51 participants available for further analysis. The diagnostic breakdown of these participants were as follows: 16 demented (10 possible Alzheimer's disease, three probable Alzheimer's disease, and three unlikely Alzheimer's disease) and 35 CIND patients. Due to the small number of dementia patients, all 16 were treated as a single group for the purpose of analysis. Table 2 contains demographic, educational, and clinical information for the CIND and dementia groups. The two groups did not differ significantly in terms of age or number of years of education (statistics shown in Table 2). There was a relatively even distribution of participants within each age decade for both groups with the following exceptions: the oldest and youngest decades for the CIND group contained few participants, as did the youngest decade for the dementia group. The majority of participants were male, and English was the predominant first language in both groups. Approximately 75% of all participants reported graduating from high school, college or university. There were no significant differences in the total score obtained on the Geriatric Depression Scale or in the number of months since the last assessment (in those cases with one or more prior assessments) (see Table 2 for statistics). The majority of CIND and Generalized Slowing in 27 Table 2 Demographic Information, Education, Depression, Mental Status, and Diagnostic Information for CIND and Dementia Participants Variable CIND Dementia t(49) n Age 40-49 50-59 60-69 70-79 80+ Average age in years Sex Men Women English as first language Yes No Education 0-8 years High school College/university 35 2(5.7) 10 (28.6) 10 (28.6) 12 (34.3) 1 (2.9) 64.3 23 (65.7) 12 (34.3) 25 (71.4) 10 (28.6) 1 (2.9) 13 (37.1) 13(37.1) 16 1 (6.3) 4 (25.0) 3 (18.8) 5(31.3) 3 (18.8) 69.0 12 (75) 4(25) 12 (75) 4(25) 1 (6.3) 8 (50.0) 5(31.3) 1.46 Generalized Slowing in 28 Post-graduate Overall years of education Depression* • Not depressed Mildly depressed Severely depressed Total score (/30) Mental status Mini Mental State Exam (/30) Previous neuropsychological assessments 0 prior assessments 1 prior assessment 2 prior assessments Number of months since last assessment Diagnostic information Age at symptom onset Age at diagnosis Global Deterioration Scale Functional Rating Scale Memory Social/Occupational Home and Hobbies 8 (22.9) 14.3 23 (65.7) 10 (28.6) 2 (5.7) 8.2 27.8 15(42.9) 20 (57.1) 0 (0.0) 16.4 59.2 63.3 2.7 2.6 2.2 1.7 2(12.5) 13.3 9 (56.3) 5(31.3) 1 (6.3) 8.5 25.7 8 (50.0) 6 (37.5) 2(12.5) 15.1 64.2 67.8 4.2 3.6 3.1 2.6 .91 .17 .36 1.39 1.30 Generalized Slowing in 29 Personal Care 1.1 1.5 Language 2.0 2.8 Problem Solving/Reasoning 2.2 3.0 Affect 2.7 3.4 Orientation 1.2 1.7 Total score (/40) 15.8 21.8 Note: Values in parentheses are percentages. * Geriatric Depression Scale (Brink, Yesavage, Lum, Heersema, Adey & Rose, 1982; Yesavage, Brink, Rose & Adey, 1986) • Missing data from one dementia patient on this test Generalized Slowing in 30 dementia participants were classified as not depressed with relatively similar proportions of participants from each group falling into the mildly and severely depressed categories. The neuropsychological assessment carried out for this study was the first for approximately half of both the dementia and CIND participants. Of those participants who had received prior assessments, there was a period of at least 11 months between the current and prior assessment for all but one participant. The single exception was one CIND participant who, due to a scheduling error, was reassessed after a period of only eight months. As expected, the dementia group scored lower on the Mini Mental State Exam and were rated as being more impaired on both the Functional Rating Scale and Global Deterioration Scale than the CIND group. However, the two groups did not differ significantly in terms of the age at which they first noticed their symptoms or the age at which they were first diagnosed. The conditions or reasons for which prescription medications were being used in the CIND and dementia groups are listed in Table 3. It is evident from looking at Table 3 that prescription medications were being taken for a wide variety of medical conditions in both groups. The most frequently used medications were anti-inflammatory/analgesics/muscle relaxants (45%) and medications for cardiac/vascular problems (35.3%). Of note, there were two CIND and six dementia participants who were currently taking medication (i.e., Aricept®) to help improve their cognitive functioning or to prevent further cognitive decline. Table 4 contains a breakdown of vitamin, mineral, supplement, and alternative medicine usage for both groups. The most commonly used vitamins were E (49.0%) and B (29.4?/o). Gingko biloba, an alternative medicine that is typically used for the purpose of improving cognitive functioning, was being taken by 31.4% of the overall sample. Generalized Slowing in 31 Table 3 Prescription Medication Usage by Category/Reason for CIND and Dementia Participants Category or reason for taking medication CPND Dementia Anti-inflammatory/Analgesic/Muscle relaxant 15 8 Cardiac/Vascular 11 7 Antidepressant 6 3 Cognitive impairment 2 6 Asthma/Allergies 4 4 Hypnotic 5 0 Diabetes 3 2 Thyroid hormone replacement 4 1 Anxiolytic/Antianxiety 2 2 Female hormonal replacement 4 0 Antiemetic 2 1 Stomach ulcer 1 2 Osteoporosis/Bone defect 2 0 Antibiotic 2 0 Pancreatic enzyme replacement 0 1 Glaucoma 0 1 Seizures 0 1 Leg cramps 1 0 Dermatitis/psoriasis 0 1 Stimulants Laxative Cirrhosis of the liver Generalized Slowing in 32 1 0 1 0 1 0 Note: Values represent the number of participants for that category in each group. Generalized Slowing in 33 Table 4 Vitamin, Mineral, Supplements, and Alternative Medicine Usage for CIND and Dementia Participants Category CIND Dementia Vitamins, minerals and supplements Vitamin E 16 9 Vitamin B 11 4 Vitamin C 6 3 Multivitamin 5 2 Calcium 3 1 Vitamin D 2 1 Vitamin A 2 0 Folic acid 1 1 Zinc 1 0 Lecithin 0 1 ternative medicines Gingko biloba 9 7 Liver oil 1 2 Garlic 1 2 Hawthorne 0 1 Mistletoe 0 1 Siberian ginseng 0 1 Generalized Slowing in 34 Primrose oil 1 0 Saw palmetto 1 0 Grape seed extract 1 0 Note: Values represent the number of participants for that category in each group. Generalized Slowing in 35 Instruments The data for the present study were obtained during the neuropsychological component of the clinic assessment. During this assessment, each patient completed a comprehensive neuropsychological test battery, covering a wide range of cognitive domains (e.g., attention, memory, language, visuospatial abilities, and executive functioning). A list of the tests that composed this test battery is provided in Appendix B. However, given that the purpose of this study was to assess the ability of performance on measures of processing speed to mediate the relationship between age and high-level cognition in CIND and dementia patients, only those tests that were deemed relevant to this purpose are described here. The tests of interest from the larger neuropsychological test battery included the Finger Tapping Test (Spreen & Strauss, 1998), the California Verbal Learning Test (Delis, Kramer, Kaplan, & Ober, 1987), and the Rey-Osterrieth Complex Figure (Rey, 1941; Osterrieth, 1944 both translated by Corwin & Bylsma, 1993). The Finger Tapping Test was relevant to this investigation as it provided a means of assessing the motor component of processing speed. High-level cognition was assessed using a measure of verbal episodic memory (the California Verbal Learning Test) and nonverbal episodic memory (the Rey-Osterrieth Complex Figure). Two additional processing speed instruments were added to the neuropsychological test battery to measure processing rate and processing capacity. Motor speed was measured by the Finger Tapping Test (Spreen & Strauss, 1998). This test requires the participants to tap a key, mounted on a manual tapper, as fast as they could using the index finger of their preferred or dominant and non-dominant hand on separate blocks of trials. The test was administered according to the instructions provided in Spreen and Strauss (1998) with the exception that five consecutively administered trials were Generalized Slowing in 36 used to obtain the average number of taps for each hand instead of giving as many trials as needed to ensure that the number of taps on each of the five trials for each hand were within five points of one another. As reported in Spreen and Strauss (1998), the test-retest reliability of this test is respectable, falling in the .58 to .90 range. There are a number of factors could that result in some of the lower reliabilities (i.e., those below .70) that have been reported with this test (Mitrushina, Boone, & D'Elia, 1999; Spreen & Strauss, 1998). Several of these factors pertain to the manner in which the test is administered. For example, some investigators employ a criterion that all five trials for each hand must be within five points of one another. This is done to avoid the influence of those trials on which a very low or high number of taps were obtained on the overall average; however, this also adds to the total time required to complete the test. Fatigue may be a factor if adequate breaks are not given in between trials. Proper test administration procedures require that the hand of the participant does not move with the finger while tapping. However, in reality, this requirement is very difficult to achieve, even after repeated instruction and requesting the participant to hold their hand down. Some participants, especially older ones, may also experience a mild degree of arthritis in their hands, but may fail to indicate this to the investigator. Processing rate was assessed via a computerized simple reaction time task. The target was the capital letter X , displayed in 2 8-point Helvetica font in the middle of the computer display of an NEC notebook computer. The participants were instructed to press the down arrow key on the keyboard as quickly as they could in response to the target. Each key press, including the first one to begin the task, elicited the next stimulus, which appeared after a randomly determined interval of either 500, 750, 1250, or 1500 ms. The participants Generalized Slowing in 37 were instructed on how to perform the task and were asked to place the index finger of their dominant or preferred hand upon the down arrow key of the keyboard. One or more brief practice trials were administered in order to familiarize the participants with the task. To begin the task, the participants pressed the down arrow key on the keyboard. Al l participants completed three blocks of 25 trials each. Each block was followed by a brief pause, the length of which was determined by the participant. The test-retest reliability of this instrument is respectable, falling in the .72 to .81 range (Graf & Uttl, 1995). Processing capacity was determined through the use of a computerized card sorting task that was modified from Rabbitt (1965). The computer display was set up to resemble a playing card that was approximately 7 cm x 7 cm in size and divided into nine squares of equal size (note: no lines corresponding to the square boundaries actually appeared on the screen). The target, which appeared randomly in one of the nine squares of the computer display, was the capital letter A on half of the trials and the capital letter B on the other half. On each trial, one of the two target letters appeared in the presence of either zero, four, or eight other letters that served as distractors. Thus, this task comprised three conditions, each of which was defined on the basis of the number of distractor letters present, (i.e., zero, four, and eight). For each of the three conditions, the distractor letters were selected randomly and without replacement from the alphabet. Both target and distractor letters were capitals and displayed in 28-point Helvetica font. The participants were instructed to sort the cards by pressing the left arrow key whenever a letter A appeared in the computer display and the right arrow key whenever the letter B appeared in the computer display. Immediately after pressing one of the arrow keys, the next card appeared on the computer screen. Prior to beginning the task, the participants were instructed how to perform the task and were asked Generalized Slowing in 38 to place their index and ring fingers of their dominant or preferred hand upon the left and right arrow keys of the keyboard, respectively. In order to familiarize the participants with the task, one or more brief practice trials were administered. When the participants understood the task, they proceeded to complete three blocks of 54 trials each. Within each block, there was an equal number of trials involving zero, four, and eight distractor letters (3 x 18 = 54 trials). Each block was followed by a brief pause, the length of which was determined by the participant. In addition to the decision time for each card, the number of errors in each of the three distractor conditions was recorded. Good test-retest reliabilities (.87 to .90) have been demonstrated with this instrument (Graf & Uttl, 1995). Episodic memory for verbal material was indexed by performance on the California Verbal Learning Test or CVLT (Delis, Kramer, Kaplan, & Ober, 1987). For the first part of this task, a list of 16 words is read to the participants. The words on this list (List A) can be divided into four conceptual categories (i.e., clothing, spices and herbs, tools, and fruits). List A is read to each participant five times, with a free-recall trial occurring immediately after each of the five presentations. A distractor or interference list (List B) of 16 words is then read to each participant. The words on List B can also be divided into four conceptual categories (i.e., fish, appliances, spices and herbs, and fruit). The tools and spices and herbs on Lists A and B are not the same items. List B is followed by a free-recall trial for the words on this list. Immediately following the free-recall trial for the words on List B, there are free- and cued-recall trials for the words on List A (Short Delay). The cues that are given are the superordinate categories for each of the items (i.e., clothing, spices and herbs, tools, and fruits). There are free- and cued-recall trials and a recognition trial for the words on List A (Long Delay) 20 minutes after the Short Delay cued-recall trial. There is also a Generalized Slowing in 39 Discriminability index that takes the proportion of false positives and misses from the recognition list into account to determine the ability of the participant to discriminate which words were from List A and which words were not (see Lezak, 1995). The CVLT was administered in accordance with the instructions provided in the test manual (Delis, Kramer, Kaplan, & Ober, 1987). Episodic memory for nonverbal material was assessed with the Rey-Osterrieth Complex Figure test or ROCF (Rey, 1941; Osterrieth, 1944 both translated by Corwin & Bylsma, 1993). For this task, the participant is given a piece of paper containing a complex two-dimensional figure and is asked to copy it as closely as they can on a separate piece of paper. When finished, or after six minutes have passed, the figure and their drawing are removed and the participant is asked to draw as much of the figure as they can remember (immediate Recall trial). The participant is then asked to draw the figure again after a 30-minute delay (Delayed Recall trial). This test was scored using a standardized set of criteria, described in Lezak (1995). Briefly, the figure is broken down into 18 separate components, and points are assigned for the presence of each of these components. Partial points are given for distorted or incorrectly placed components. The ROCF was administered in accordance with the instructions published in Lezak (1995). Although the ROCF was administered to 33% of the sample by the clinic psychometrist, the principle investigator rescored the tests that were given to these participants. The inter-rater reliabilities obtained from the twice-scored test protocols were quite high: .92, .99, and .99 for the copy, immediate, and recall scores, respectively. These values are consistent with published inter-rater reliabilities (Lezak, 1995; Mitrushina, Boone, &D'Elia, 1999). Generalized Slowing in 40 Procedure The data for the present study were obtained during the neuropsychology component of the clinic assessment. Al l patients were assessed in a small and quite room located in the basement of the extended care facility of the hospital. A total of approximately 4.5 hours was required to complete the clinical neuropsychological assessment. A mandatory lunch break was given in the middle of the test battery with extra breaks given as required. Although the order of tests given in the clinical neuropsychological assessment was not fixed, the ROCF was administered near the beginning of the assessment. The finger tapping test was usually given in the middle of the assessment. The CVLT was administered approximately three quarters into the test battery. The two processing speed tests that were added onto the test battery, simple reaction and card sorting, were always administered at the end of the assessment with the former always being given before the latter. Data Preparation and Analysis Data Preparation. The data obtained from all test instruments were screened for outliers, defined as scores that were plus-or-minus three standard deviations away from the respective mean. Al l identified outliers were replaced with the value represented by the third standard deviation away from the respective mean. In all instances, this resulted in replacement of fewer than 4% of the scores. This treatment of outliers was preferred because of the small sample sizes. The CVLT, ROCF, and Card Sorting Accuracy data were also examined for the presence of floor and ceiling effects, which were defined as zero and perfect scores by more than 20% of the participants, respectively. Floor effects were only present in the performance of the dementia group on the Short and Long Delay free-recall trials of the CVLT. Ceiling effects were observed in the performance of the CIND group on the number Generalized Slowing in 41 of Hits during the recognition trial of the CVLT and in the accuracy levels for all three Card Sorting distractor conditions; the performance of the dementia group was at ceiling for the accuracy levels for the four distractor Card Sorting condition. Regarding missing data, there was only one participant, from the dementia group, who did not complete the finger tapping test for the nondominant hand. His value was replaced with the average of the dementia group. Data Analysis. Al l analyses were performed using the Statistical Package for the Social Sciences (SPSS), Windows Version 8.0.0. The data were analyzed using two strategies. The first strategy involved comparing the performance of the dementia and CIND groups on the measures obtained from each of the test instruments. To accomplish this, a series of independent sample t-tests were performed. The processing speed measures that were analyzed included the mean number of taps for the dominant and nondominant hands, mean simple reaction time obtained across all three blocks of 25 trials, and mean card sorting reaction times and accuracy levels obtained from each distractor condition. Regarding the CVLT, the performance across the first five trials of List A was analyzed using a 2 x 5 ANOVA with group as the between subjects factor and trial as the within subjects factor. A 2 x 2 ANOVA with group as the between subjects factor and recall type (free vs. cued) as the within subjects factor was conducted for the Short Delay free- and cued-recall trials. A similar ANOVA was conducted for the Long Delay free- and cued-recall trials. Independent sample t-tests were used to compare the performance of both groups on the mean number of words correctly recalled from the distractor list (List B), the mean number of words from List A that were correctly recognized after the Long Delay (Recognition Hits) and the Discriminability index from the recognition trial. The Generalized Slowing in 42 performance of the CIND and dementia groups on the ROCF copy, immediate, and delayed recall scores were also compared using independent sample t-tests. The degrees of freedom were adjusted for all comparisons involving independent sample t-tests when the homogeneity of variance assumption was violated (the method used by SPSS is the Welch formula). The Ffuynh-Feldt (1976) correction for degrees of freedom was used for all ANOVAs in which assumption of sphericity was violated. To reduce the probability of committing a Type I error, the alpha level was reduced for all comparisons involving independent sample t-tests according to the Bonferroni correction procedure (.05/15 comparisons = .003). The alpha level was fixed at .05 for each of the three ANOVAs. To provide some degree of how the performance of the CIND and dementia groups would compare to the performance of healthy adults of similar age, standardized scores were computed using normative data for each test instrument. For the Finger Tapping Test, Simple Reaction Task, and Card Sorting Task, the means and standard deviations reported in Graf and Uttl (1995) were used to obtain z-scores for each participant. The performance on the CVLT and the ROCF, were standardized using the normative data contained in Paolo, Troster, and Ryan (1997) and Meyers and Meyers (1995), respectively. The purpose of the second strategy was to test the hypotheses of this investigation. To accomplish this goal, a series of hierarchical regression analyses were performed. These analyses involved determining the ability of age to predict performance on the CVLT and ROCF before and after statistically controlling for the influence due to processing speed measures in both the dementia and CIND groups. For all hierarchical regression analyses, finger tapping speed was indexed by the average tapping performance for the dominant or Generalized Slowing in 43 preferred hand. Simple reaction was defined as the average reaction time obtained across the three blocks of 25 trials. The index of card sorting was the average time taken to correctly sort cards in the four distractor condition. In addition to the processing speed predictor variables, several covariate variables were also entered into the hierarchical regression analyses. These included the following: sex, number of years of education, number of medications (total number of prescription drugs, vitamins, minerals and supplements, and alternative medicines), depression (total score on the Geriatric Depression Scale), and the copy score on the ROCF. The first four of these variables have all been shown to influence performance on many neuropsychological tests (Lezak, 1995; Vernon-Wilkinson & Tuokko, 1990). The purpose for using the copy score as a covariate was to remove the variance associated with the drawing component involved in the ROCF, thereby obtaining a more valid index of the memory component of this task. In other words, it was assumed that the ability to recall the ROCF would be influenced by the original ability to copy the figure, with higher copy scores being related to higher recall scores. For example, if a participant only copied a third of the details of the figure, they would be less likely to remember more than these details on subsequent recall. For each hierarchical regression analysis, the variance explained by age alone was determined first. The next part of the analysis involved entering each covariate alone. Two models were then tested. The first model, which involved the processing speed measures and age, was similar to the one used by Graf and Uttl (1995) except that they used a different measure of verbal episodic memory, and they did not use a measure of nonverbal memory. This allowed for a qualitative comparison of the results obtained between the sample of healthy adults tested by Graf and Uttl (1995) and those from the CIND and dementia groups Generalized Slowing in in this investigation on these processing speed measures. The final model contained only those predictor variables that were found to be significant in the preceding analyses Generalized Slowing in 45 CHAPTER FOUR Results The present study is still in progress, and is intended to^examine the influence of the Generalized Slowing Hypothesis in dementia and CJND patients. A power analysis conducted prior to this investigation indicated that with 60 participants in each group, the power to detect a significant effect due to age would be .75 (Cohen, 1988). This thesis is a preliminary report of the results obtained from 16 dementia and 35 CIND patients that have been run in the investigation to date. The recruitment of participants for this project is still underway. More concrete and reliable analyses and conclusions from these analyses will need to wait until larger sample sizes are obtained. The analyses used to examine the Generalized Slowing Hypothesis in the present investigation were guided by those conducted byGrafandUttl(1995). Performance on Processing Speed and High-Level Cognitive Test Instruments Finger Tapping. This test required the participants to tap the index finger of both their preferred or dominant and nondominant hands as fast as they could for 10 seconds. Five consecutive trials were completed for each hand and the dependent variable of interest was the average number of finger taps. The mean number of finger taps for each hand in both the CIND and dementia groups is reported in Table 5. As can be seen, the mean values for both hands are higher in the CIND group than in the dementia group; however, neither of these differences were significant (see Table 5 for statistics). It can also be seen from Table 6 that the average performance of the CPND participants was around one standard deviation below normal. The mean performance of the dementia participants was closer to 1.5 standard deviations below normal. Generalized Slowing in 4 Table 5 Means and Standard Deviations (in parentheses) for CIND and Dementia Participants on Processing Speed Tasks Instrument/Measure CIND Dementia t(49) Finger Tapping Test (Number of taps) Dominant hand Nondominant hand Simple Reaction Mean reaction time (ms) Card Sorting with 0 distractors Mean accuracy Mean reaction time (ms) Card Sorting with 4 distractors Mean accuracy Mean reaction time (ms) 43.09 (6.81) 39.27 (6.26) 356.30 (49.52) .98 (.03) 766.84 (152.70) .97 (.03) 1014.69 40.84 (9.59) 37.92* (7.57) 399.71 (66.74) .95 (.07) 915.51 (214.10) .93 (.11) 1311.22 .85* .67 2.60 2.16 2.91 1.12. 3.15* (280.65) (390.02) Generalized Slowing in 47 Card Sorting with 8 distractors Mean accuracy Mean reaction time (ms) .97 (.03) 1324.31 .93 (.09) 1717.90 1.84* 2.52 (453.60) (550.39) Note: In cases of unequal variances, the degrees of freedom were adjusted using Welch's formula: •df=22, v df = 16 A df = 17 * One participant in this group was not administered this test * p < .003 (in accordance with the Bonferroni correction) Generalized Slowing in 48 Table 6 Mean Standardized Scores of CIND and Dementia Participants Relative to Normative Data for Each Test Instrument Instrument/Measure CIND Dementia Finger Tapping Test (Number of taps)* Dominant hand -1.05 -1.39 Nondominant hand -96 -1.76 Simple Reaction (ms)+ Mean reaction time Card Sorting (ms) + 0 Distractors 4 Distractors 8 Distractors CVLT (Number of words correctly recalled/recognized)* Average of trials 1 to 5 (List A) Trial 1 Trial 5 ListB Short delay free-recall Short delay cued-recall Long delay free-recall Long delay cued-recall -1.12 -2.09 -.24 -1.74 -.46 -2.74 -.32 -2.48 -.15 -1.87 +.19 -.87 -.25 -1.90 -.22 -.88 -.42 -1.91 -.19 -2.09 -.22 -2.01 -.13 -2.18 Generalized Slowing in 49 Recognition (number of hits) +.06 -.86 Discriminability Index -.22 -2.74 ROCF test score* Copy above -1.00 standard deviation 69 % 69 % -1.00 to -1.99 standard deviations 17 % 13 % greater than -2.00 standard deviations Delayed recall 14% 18% Immediate recall +-02 -1.47 .19 -1.46 Note: Standardized scores were obtained by computing z-scores using the following sources of normative data: * Graf and Uttl (1995), v Paolo, Troster, and Ryan (1997), * Meyers and Meyers (1995) Generalized Slowing in 50 Simple Reaction. During this computerized task, participants were required to respond as quickly as possible to the presentation of a simple stimulus (i.e., an X). The average reaction time for each participant was obtained across three blocks of 25 trials. The mean reaction times for both CIND and dementia groups are presented in Table 5. Although, the mean reaction time was faster in the CIND group than in the dementia group, this difference was not significant (see Table 5 for statistics). The standardized scores reported in Table 6 reveal that the average performance of the CIND and dementia participants were approximately one and two standard deviations below normative standards, respectively. Card Sorting. This computerized task involved sorting cards on the basis of whether they contained an A or B on them. There were three conditions with zero, four, or eight distractor letters present on each card. The critical dependent variables in each of these three distractor conditions were the accuracy of performance and average time taken to sort the cards. The accuracy and reaction time data obtained from the CIND and dementia groups for each of the three conditions of this task are reported in Table 5. As can be seen in Table 5, the participants were very accurate, with the mean level of accuracy being 93% or greater in all conditions for both groups. However, the accuracy of performance in the dementia group was more variable than in the CIND group. In addition, the performance of the CIND group was at ceiling for all three conditions, and the performance of the dementia group was at ceiling for the four distractor condition only. Therefore, it is not surprising that the two groups did not differ significantly in terms of their level of accuracy. Only those trials involving a correct response were used to obtain the average reaction times for this task. It is apparent from looking at Table 5 that the CIND participants were faster than the dementia participants in all three distractor conditions. However, only the Generalized Slowing in 51 difference between the two groups in the condition involving four distractor letters was significant (statistics reported in Table 5). As reported in Table 6, the average performance of the CIND participants on each of the three distractor conditions was less than one half a standard deviation below normal. In contrast, the average performance of the dementia participants was approximately 1.5 standard deviations below normal. CVLT. The dependent variables of interest were the average number of words correctly recalled on the seven free-recall and two cued-recall trials for List A, the free-recall trial for List B, the average number of words from List A that were recognized after the 20 minute delay (Recognition Hits), and the Discriminability index for the recognition trial. The data obtained from the CIND and dementia groups for each of these variables are presented in Figure 1. To examine the performance on the first five trials of List A, a 2 x 5 ANOVA was conducted with group as the between subjects factor and trial as the within subjects factor. There were significant main effects for both group, F(l , 49) = 22.02, MSE = 675.99, and trial, F(4,49) = 30.81, MSE = 68.64. The interaction between trial and group was also significant, F(4, 49) = 5.53, MSE = 12.32. To determine whether the ability of each group to learn across repeated trials was significant and whether or not this ability followed a linear or nonlinear pattern, one-way ANOVAs with subsequent trend analyses were performed for each group. The one-way ANOVAs were significant in the CIND group, F(4, 34) = 45.98, MSE = 97.39, and in the dementia group, F(4, 15) = 5.13, MSE = 9.66. The linear trends were also significant for both the CIND group, F(l, 34) = 126.26, MSE = 370.29, and the dementia group, F( l , 15) = 7.89, MSE = 29.76. Regarding the nonlinear trends, the quadratic contrast for the CIND group was significant, F(l , 34) = 6.62, MSE = 18.80, and the fourth Generalized Slowing in 52 CIND Dementia A • + + + A l A2 A3 A4 A5 Bl Trials Trial List A List B SD SD LD LD FR CR FR CR Trials List A REC Figure 1. The mean performance of the CIND and dementia groups on the various CVLT measures. Areas that are enclosed by boxes indicate performance floor or ceiling effects, in which more than 20 percent of the participants obtained zero or perfect scores respectively. Note: SD FR = Short delay free-recall, SD CR = Short delay cued-recall, LD FR = Long delay free-recall, LD CR = Long delay cued-recall, REC = Recognition Hits score (number of true positives) Generalized Slowing in 53 order polynomial was significant for the dementia group, F(l , 15) = 5.96, MSE = 7.23. Thus, both groups demonstrated significant learning ability. However, this ability to learn was not simply a linear increase in the number of words correctly recalled across repeated trials. Free-recall of the words from the distractor list (List B) was significantly higher in the CIND group than the dementia group, t(49) = 2.46. The results of the Short Delay free- and cued-recall trials were subjected to a 2 x 2 ANOVA, with group as the between subjects factor and recall type (free and cued) as the within subjects factor. Significant main effects were present for both group, F(l , 49) = 19.13, MSE = 548.63, and recall type, F(l , 49) = 35.11, MSE = 69.47. The interaction between group and recall type was not significant, F(l, 49) = .86, MSE = 1.70. A similar ANOVA was conducted for the Long Delay free- and cued-recall trials. Again, there were significant main effects for group, F(l , 49) = 20.10, MSE = 689.81, and recall type, F(l, 49) = 21.77, MSE = 26.57. There was no significant interaction between group and recall type, F(l , 49) = .18, MSE = .22. Regarding the recognition trial conducted after the 20 minute delay, the difference between the CIND and dementia groups in terms of Recognition Hits was not significant, t(49) = 1.65. Although not shown in Figure 1, the difference between the CIND and dementia patients to discriminate List A words from non List A words on recognition was significantly different, t(49) = 4.31, with the CIND group (M = 89.43%; SD = 9.39) outperforming the dementia group (M = 74.87%; SD = 14.47). This latter result suggests that although the dementia participants said "yes" to most of the words from List A during the recognition trial, they also said "yes" to more words from List B and words that were from neither List A nor B. The results obtained from the short and long delay trials of the CVLT should be interpreted with some degree of caution due to the presence of floor and ceiling effects in Generalized Slowing in 54 these measures. In Figure 1, performance floor and ceiling effects are indicated by the boxes surrounding the plotted data points that are affected. The performance on the free-recall trials at both short and long delays were at the floor in the dementia group. The performance of the CIND group on the number of Recognition Hits was at ceiling. There were no floor or ceiling effects present in the first five free recall trials of List A or in the recall trial of List B. To provide a more reliable estimate of learning and memory ability, the average number of words correctly recalled across the first five trials of List A was chosen to be used in subsequent analyses (i.e., hierarchical regression analyses) rather than the single recall trial of List B. The means and standard deviations for the performance of the CIND and dementia groups across the first five trials were 8.57 (SD = 2.37) and 5.06 (SD = 2.70), respectively. The standardized scores contained in Table 6 reveal that the average performance of the CIND participants was within one half of a standard deviation above and below the mean of healthy individuals. The mean performance of the dementia participants on each of the CVLT measures was within approximately 1 to 2.5 standard deviations below normal. ROCF. Participants were required to copy a complex two-dimensional figure and then reproduce this figure on immediate and delayed recall trials. The critical dependent variables were the scores obtained on the copy, immediate and delayed recall of the figure. These results are shown in Figure 2. As can be seen, the copy performance for both the CIND (M = 31.33; SD = 4.25) and dementia groups (M = 30.12; SD - 4.71) were quite high. This difference was not statistically significant, t(49) = .91. On the immediate recall of the figure, the average performance of the CIND group (M = 15.90; SD = 7.54) was significantly higher than the dementia group (M = 7.44; SD - 5.87), t(49) = 3.95. Similarly, on the Generalized Slowing in 55 36 T Copy Immediate Delay Test Trial Figure 2. The mean performance of the CIND and dementia groups on the various ROCF measures. Generalized Slowing in 56 delayed recall of the figure, the average performance of the CIND group (M = 15.00; SD = 7.42) was significantly higher than the dementia group (M = 7.44; SD = 6.22), t(49) = 3.54. Referring to the standardized scores reported in Table 6, it is apparent that the average performance of the majority of CIND and dementia participants was above one standard deviation below normal. The mean performance of the CPND participants was comparable to normative standards on both the immediate and delayed recall of the figure. In contrast, the average performance of the dementia participants on the immediate and delayed recall of the figure was approximately 1.5 standard deviations below normal. Given that the average recall of the first five trials of List A on the CVLT was selected to provide a measure of verbal episodic memory for all subsequent analyses, the immediate recall of the ROCF was chosen to provide a measure of nonverbal episodic memory for all subsequent analyses. Hierarchical Regression Analyses Hierarchical regression analyses were conducted to determine whether age is related to performance on the CVLT and ROCF before and after statistically controlling for processing speed in the CPND and dementia participants. An essential requirement in conducting these analyses is that age be significantly correlated with test performance on the dependent variable of interest. Table 7 contains the Pearson product-moment correlation coefficients and coefficients of determination (r2) for the relationships between age and performance on the CVLT and ROCF in the CIND and dementia groups. Al l correlations between age and test performance were significant except for the performance of the dementia group on the ROCF, in which age only accounted for 3% of the variance. A preliminary analysis revealed that age was not a significant predictor of ROCF performance Generalized Slowing in 57 Table 7 Correlations and Coefficients of Determination (in parentheses) Between Age and Performance on CVLT and ROCF for CIND and Dementia Participants CVLT ROCF CIND -.67*** -.37* (.45) (.14) Dementia -.55*** -.16 (.30) (.03) * p<.05, *** p<.001 Generalized Slowing in 58 in the dementia group, F(l , 14) = .38. Therefore, the performance of the dementia group on the ROCF was not examined further in these analyses. When performing these analyses, one also assumes that the relationship between the variables is linear. Therefore, the relationships between age and performance on the CVLT and ROCF were examined for the presence of significant nonlinear components. This was accomplished using regression analyses in which age was entered first into the equation predicting performance on the CVLT and ROCF (Cohen & Cohen, 1983). The amount of variance that was explained by next entering age-squared was determined. If this latter value was significant (i.e., p_ <.05), the relationship contained a significant degree of nonlinearity. The nonlinear components of the relationship between age and performance on the CVLT and ROCF were not significant in the CIND group, F(l, 33) = 3.93, and F(L 33) =1.41, respectively. However, the nonlinear component of the relationship between age and CVLT performance in the dementia group was significant, F(l, 13) = 6.12. A scatterplot of the CVLT test scores as a function of age in the dementia group is graphed in Figure 3. As can seen, there are several data points that are likely contributing to the significance of the nonlinear aspects of the relationship between age and CVLT performance in this group. The nonlinear component of this relationship remained even after performing logarithmic and square root transformations of the age variable. Given that this investigation was exploratory in nature, I decided to proceed as if the nonlinear component of this relationship was not significant, with the full recognition that the tenability of the conclusions that could be drawn from the analysis in the dementia group were compromised. Preliminary analyses revealed that in both groups, finger tapping was not a significant predictor of performance on either measure (CVLT and ROCF). The amount of variance Generalized Slowing in 59 30 40 50 60 Age 70 80 90 Figure 3. Scatterplot of the CVLT test scores as a function of age in the dementia group. Note the presence of the data points (encircled) that are contributing to the nonlinear component of the relationship between age and CVLT performance. Generalized Slowing in 60 accounted for by finger tapping performance was less than 5% in all cases. Therefore, the finger tapping measure was not included in any of the hierarchical regression analyses. CVLT. The results of the hierarchical regression analyses for the CIND group are reported in Table 8. As can be seen, age was a significant predictor of performance on the CVLT (see Tables 8 for statistics). Regarding the ability of the covariates to predict CVLT performance, only the variance accounted for by sex was significant. In the first model, which examined the ability of the processing speed components and age to predict performance on the CVLT, all three predictors explained unique and significant amounts of variance in CVLT performance. When the total variance accounted for by both measures of processing speed (simple reaction and card sorting) were taken into account, there was an attenuation of 62% ([45-17J/45 x 100 = 62) in the predictive ability of age to explain CVLT performance. The final model, which tested only those predictors that were significant in prior analyses included sex, simple reaction, card sorting, and age. Each of these variables contributed a significant amount of unique variance in CVLT performance. The predictive ability of age was reduced by 87%) ([45-6J/45 x 100 = 87) after the control of these variables. The results of the hierarchical regression analyses for predictor variables on the CVLT in the dementia group are reported in Table 9. Again, age was a significant predictor of performance on the CVLT (see Table 9 for statistics). Not one of the covariate variables were significant predictors of CVLT performance. Similarly, not one of the predictors in the first model were significant. Therefore, the only significant predictor of CVLT performance in the dementia participants was age. Although none of the processing speed measures were significant predictors, the magnitude of attenuation in the predictive ability of age to explain CVLT performance after Generalized Slowing in 61 Table 8 Summary of Hierarchical Regression Analyses for Variables Predicting Performance on the California Verbal Learning Test in CIND Participants (N = 35) Change Statistics Predictors R 2* A R 2 F df Age only 1. age .45 26.43*** 1,33 Covariates alone 1. sex .20 7.98** 1,33 Model Testing I) Processing Speed Components and Age 1. simple reaction .20 8.15** 1,33 2. simple reaction, card sorting .38 .18 9.16** 1,32 3. simple reaction, card sorting, age .55 .17 11.61** 1,31 II) Final Model 1. sex .20 7.98** 1,33 2. sex, simple reaction .47 .27 16.61*** 1,32 3. sex, simple reaction, card sorting .54 .07 4.39* 1, 31 4. sex, simple reaction, card sorting, age .60 .06 4.47* 1, 30 * Proportion of variance explained in the CVLT * p<.05, ** p<.01, *** p<.001 Generalized Slowing in 62 Table 9 Summary of Hierarchical Regression Analyses for Variables Predicting Performance on the California Verbal Learning Test in Dementia Participants (N = 16) Change Statistics Predictors R2+ A R 2 F df Age only 1. age .30 6.10* 1,14 Model Testing I) Processing Speed Components and Age simple reaction .09 1.46 1,14 simple reaction, card sorting .10 .01 .12 1,13 simple reaction, card sorting, age .32 .22 3.82 1,12 * Proportion of variance explained in the CVLT * p<.05 Generalized Slowing in 63 statistically controlling for the measures of processing speed was determined. This allowed for the qualitative comparison of attenuation magnitudes for the CIND and dementia groups. After partialling out the influence due to processing speed, the predictive ability of age was reduced by only 27% ([30-22J/30 x 100 = 27). Therefore, these results suggest that the role of processing speed in mediating the relationship between age and CVLT performance was more pronounced in the CIND group than in the dementia group. ROCF. Table 10 contains the results of the hierarchical regression analyses for the CIND group. Age was a significant predictor of performance on the ROCF (see Table 10 for statistics). Regarding the covariates, the copy score and medication variables were the only significant predictors of ROCF performance. Simple reaction was the only predictor in the first model that explained a significant unique amount of variance in performance on the ROCF. In the final model, the copy score was the only the significant predictor of ROCF performance. Despite the fact that neither the card sorting measures nor age were significant predictors of ROCF performance in the first model, the magnitude of attenuation in the ability of age to predict ROCF performance after statistically controlling for the measures of processing speed was determined. This allowed for the qualitative comparison of the attenuation magnitudes observed for the CVLT and the ROCF measures in the CIND group. The statistical control of the processing speed measures resulted in a 43% reduction ([14-8]/14 x 100 = 43) in the predictive ability of age to explain ROCF performance. Therefore, these results suggest that the role of processing speed in mediating the relationship between age and CVLT performance was more pronounced than for ROCF performance in the CIND Generalized Slowing in 64 Table 10 Summary of Hierarchical Regression Analyses for Variables Predicting Performance on the Rey-Osterrieth Complex Figure in CIND Participants fN = 35) Change Statistics Predictors R 2* A R 2 F df Age only 1. age Covariates only .14 5.38* 1,33 1. copy score .30 14.40** 1,33 2. medications .22 9.44** 1,33 Model Testing I) Processing Speed Components and Age 1. simple reaction .22 9.60** 1,33 2. simple reaction, card sorting .23 .01 .18 1,32 3. simple reaction, card sorting, age .31 .08 3.76 1,31 II) Final Model 1. copy score .30 14.40** 1,33 2. copy score, medications .38 .08 4.04 1,32 3. copy score, medications, simple reaction .41 .03 1.52 1,31 4. copy score, medications, simple reaction, age .44 .03 1.40 1,30 * Proportion of variance explained in the ROCF * p<.05, ** p<.01 Generalized Slowing in group. The final model was treated in a similar manner as the first model. After the statistical control of all of three variables (copy score, medications, and simple reaction), there was an attenuation of 79% ([14-3J/14 x 100 = 79) in the ability of age to predict performance on the ROCF. Generalized Slowing in 66 CHAPTER FIVE Discussion The purpose of this research was to determine whether the Generalized Slowing Hypothesis could explain the cognitive impairments observed in dementia and CIND patients. There were two hypotheses in this investigation. The first hypothesis was that the ability of age to predict performance on the CVLT and ROCF after partialling out the influence due to measures of processing speed would be attenuated by at least 60%. Results of the hierarchical regression analyses demonstrated attenuations of 62, 43, and 27% in the predictive ability of age to explain performance on the CVLT and ROCF in the CIND group and the CVLT in the dementia group, respectively. The magnitude of attenuation in the ability of age to predict ROCF performance in the dementia group was not determined due to the fact that initially, age only accounted for 3% of the variance in this measure, leaving essentially no predictive ability to attenuate. Thus, the first hypothesis was supported for only the CVLT performance in the CIND group. The second hypothesis was that attenuations in the predictive ability of age observed after the statistical control of processing speed would be similar for the CVLT and ROCF. The rationale behind this hypothesis was that if generalized slowing is present in these two groups, then all cognitive domains should be affected equally. This hypothesis was not confirmed due to the finding of differences in the magnitude of attenuation between the CVLT and ROCF in the CIND group. Overall, the results reported in this investigation, although preliminary in nature, do not provide support for the Generalized Slowing Hypothesis in these two patient groups, as evidence in favor of generalized slowing would have required support for both hypotheses. Generalized Slowing in 67 When interpreting the results of this investigation, it is necessary to be cautious. Failure to provide support for generalized slowing in the dementia and CIND groups could be attributed to the lack of sufficient power to examine the Generalized Slowing Hypothesis in these two groups, One obvious factor that would contribute to insufficient power is the small sample sizes obtained in this investigation. This factor could explain why neither of the hypotheses were supported. An additional factor that could explain why the magnitude of attenuations were not similar for the CVLT and ROCF (the second hypothesis) in the CIND group pertains to the variability of performance on these two measures. For example, in both the CIND and dementia groups, the standard deviations for the immediate recall of the ROCF were at least twice the standard deviations for the average of the first five trials on the CVLT. Ideally, if one wishes to compare the ability of an independent variable to predict performance on two separate dependent variables, the spread of scores around the mean on those two dependent variables should be similar. The differences in terms of attenuation magnitudes using the CVLT and ROCF in the CIND group could also have been due to the age effects present on these tests. Age accounted for 45% of the variance in CVLT performance; whereas it only accounted 14% of the variance in the ROCF performance. According to the Generalized Slowing Hypothesis, age and processing speed are highly related to one another, with increases in the former being associated with decreases in the latter. As a result, performance on the CVLT may have been more sensitive to the effects of generalized slowing than performance on the ROCF. Generalized Slowing in 68 Theoretical and Practical Implications of This Research One of the important theoretical implications raised by this research involves the manner in which the Generalized Slowing Hypothesis was tested. The criterion that I used to test this hypothesis was that the attenuation in the ability of age to explain performance on the CVLT and ROCF after statistically controlling for measures of processing speed needed to be at least 60%. There were two reasons for choosing this value. The first was that attenuations that are greater than 60% have been classified as major (Salthouse, 1992a). The second reason was based on the review conducted by Salthouse (1996a) demonstrating this magnitude of attenuation to be the lowest found in 29 different studies examining generalized slowing in healthy adults. I made the assumption that if generalized slowing was present in these two groups, then one should find attenuations that are at least equal in magnitude to those observed in the normal population. However, the argument can be made that the magnitude of attenuation should actually be greater than that which is observed in healthy adults (i.e., greater than 60%). A number of investigators have adopted the hypothesis that normal and pathological cognitive aging represent ends on a continuum (Ferris & Kluger, 1996; Huppurt, et al., 1994; Rediess & Caine, 1996; Smith, Petersen, Parisi, Ivnik, Kokmen, Tangalos, & Waring, 1996). According to this view, dementia is simply an accelerated version of the normal aging process. If this is the case, and the Generalized Slowing Hypothesis is correct, there should be exaggerated slowing in these two groups compared to healthy adults. But, how does one define exaggerated slowing in these two patient groups? In other words, what degree of attenuation in the ability of age to predict performance on tests of high-level cognition after partialling out the influence due to processing speed would be Generalized Slowing in 69 required? The hierarchical regression model tested in this investigation involving the processing speed components and age was very similar to the one tested in healthy adults by Graf and Uttl (1995), with the exception that they included finger tapping performance and they used a different measure of verbal episodic memory. Their results demonstrated a 90% attenuation in the ability of age to explain memory performance after partialling out the variance accounted for by the processing speed measures. Given this finding, would a value greater than 90% be required in CIND and dementia patients or would a value closer to 100% be necessary? A problem arises in that the majority of studies examining generalized slowing in healthy adults have used large samples that cover a wide range of ages (e.g., from 20 to over 80). However, the youngest CIND and dementia patients in the present investigation were in their forties. Perhaps, it is unreasonable to expect that with a restricted age range (e.g., from 40 to over 80) the degree of generalized slowing would be greater than that observed in healthy adults across a wider range of ages. One way to examine this issue would be to test large samples of healthy adults and dementia patients (or CIND patients) that also have similar age ranges. The attentuation magnitudes could then be compared directly to see if generalized slowing is exaggerated in dementia patients relative to healthy adults. Another important theoretical implication raised by this research is that generalized slowing may not be present in these two patient groups. If future research with CIND or dementia patients consistently fails to provide support for generalized slowing, what would be the fate of the Generalized Slowing Hypothesis? The finding of one group of patients in which generalized slowing does not apply would provide evidence for a dissociation between processing speed and high-level cognition. At a minimum, this would require the Generalized Slowing Hypothesis to be modified. To illustrate this point, consider the effect Generalized Slowing in 70 that the dissociation between implicit and explicit memory has had upon existing theories of long-term memory as a single system. Implicit memory refers to the unconscious influence of prior experience on subsequent performance; whereas explicit memory refers to the conscious recollection of information (Graf & Schacter, 1985). One major source of support for this distinction came from a series of studies in the mid 1980s demonstrating that although explicit memory functioning is compromised in amnesic patients, there is preservation of implicit memory abilities (Graf & Schacter, 1985; Graf, Squire, & Mandler, 1984; Graf, Shimamura, & Squire, 1985). These findings provided support for the existence of multiple memory systems rather than a single memory system. Regarding the relationship between normal aging and dementia, a dissociation between processing speed and high-level cognition would provide evidence against the continuity view, which asserts that dementia is merely an exaggeration of the normal aging process. In other words, there are processes over and above those associated with normal aging that lead to the cognitive impairment in these two patient groups. The idea of modifying the Generalized Slowing Hypothesis is not a new one. The distinction between strong and weak versions of this hypothesis has been made (Cerrella, Poon, & Fozard, 1981). The strong version of the hypothesis, which was the one tested in the present investigation, states that all aspects of cognition that show decline are affected equally. The weaker view of this hypothesis allows for some aspects of cognition to be affected more than others. As pointed out by Johnson and Rybash (1993), the strong view of this hypothesis can be disregarded on the basis that the magnitude of age differences present in the normal elderly on a given cognitive task depends upon the type of processing that is Generalized Slowing in 71 involved. Therefore, the weaker version of the Generalized Slowing Hypothesis may need to be adopted in lieu of the stronger version. Consistent with the weaker version of the Generalized Slowing Hypothesis is the sequence of neuropathological changes that have been shown to accompany dementia. For example, there is evidence demonstrating that one of the earliest brain regions to manifest neuropathological changes in Alzheimer's disease is the entorhinal cortex and other regions in the temporal lobe (Gomez-Isla, Price, McKeel, Morris, Growdon, & Hyman, 1996; Gomez-Isla, Hollister, West, Mui, Growdon, Petersen, Parisi, & Hyman, 1997). A link has been made between these neuropathological changes and the early verbal memory impairments that are often reported in Alzheimer's disease (Albert, 1996; Damasio, Van Hoesen, & Hyman, 1995). It may be the case that these changes in the central nervous system serve to negatively affect some cognitive domains (e.g., verbal memory) while leaving others more or less unaffected. This might result in the effect of generalized slowing being present in some but not all cognitive domains, at least in the early stages of the disease process. However, one major problem associated with adopting the weaker version of this hypothesis is that it is not falsifiable. For example, for every nonsignificant finding of generalized slowing in a given cognitive domain, the investigator could simply state that the slowing in that domain is not as prominent as it is in other domains (Johnson & Rybash, 1993; Salthouse, 1985). What theoretical and practical implications would follow if the results of future investigations provided consistent support for the Generalized Slowing Hypothesis? If this were the case, the hypothesis that dementia represents an accelerated version of the normal aging process would also be supported. From a practical point of view, one would be able to Generalized Slowing in 72 examine the neurobiological and cognitive changes that accompany the normal aging process for potential clues to explain the mechanisms and processes involved in dementia. One could also use the changes that occur in dementia to provide testable hypotheses about the normal aging process. Another practical implication of generalized slowing being present in CIND and dementia patients is that the standard implementation of computerized processing speed tests would provide useful information to existing clinical neuropsychological test batteries. Given the complexity of the cognitive impairments observed in dementia, it is unlikely that any single test will identify, with certainty, which individuals will develop dementia. However, the use of processing speed measures would provide additional information for the clinican to use in determining which individuals should receive further testing. In general, computerized tests are easy and quick to administer and they tend to be more objective than paper-and-pencil tests (Anastasi, 1988; Lezak, 1995). These qualities are desirable in that they allow for the mean performance on a given measure to be based upon multiple blocks of trials rather than a single trial, as is the case for most paper-and-pencil neuropsychological tests. The effect of this would be to increase the reliability of the measure. Limitations of This Research There are several important limitations of this study. The most obvious of these is the small number of CIND and dementia participants. An additional limitation involved the administration of the tests. In the present investigation, all tests were administered on only one occasion. This may have resulted in these tests being less reliable than if they had been given more than once. A third limitation pertains to the heterogeneity of the two groups. For example, the diagnosis of CIND is applied to individuals who demonstrate cognitive Generalized Slowing in 73 impairment but who are not demented. There are, however, many potential causes of cognitive impairment (e.g., depression, prior head injury, medication use, and pre-existing medical conditions) that would result in an individual being diagnosed as CIND. In addition, the dementia group in the present investigation was composed of patients with different diagnoses (i.e., unlikely, possible, and probable Alzheimer's disease), and varying levels of symptom severity (i.e., Global Deterioration Scale scores greater than 3). These limitations could be overcome in the future by obtaining larger sample sizes and using more observations per subject. A useful strategy to employ in subsequent investigations would be to determine whether the attenuation magnitudes found for different cognitive domains (e.g., verbal vs. nonverbal memory; memory vs. problem-solving) are significantly different in the same group of CIND or dementia patients. In addition, one could also determine whether attenuation magnitudes are significantly different across various groups (e.g., CIND patients vs. healthy adults, Alzheimer's patients vs. healthy adults, CIND patients vs. Alzheimer's patients) and across different severity stages in the same condition (e.g., possible Alzheimer's patients vs. probable Alzheimer's patients). Although the present investigation determined and controlled for the influence of several covariate variables (i.e., sex, number of years of education, medication usage, depression, and the ROCF copy score), the limitations associated with the heterogeneity of these two patient groups could be overcome further by identifying and controlling for the influence of additional covariates such as prior head injury, medical conditions, motivation, anxiety, and hypervigilance. In addition, it would be interesting to conduct a longitudinal study in which generalized slowing is examined annually in CIND patients. In this manner, one could Generalized Slowing in 74 examine whether generalized slowing was more prominent on the first assessment in those CIND patients who eventually develop dementia than in those CIND patients who do not. If generalized slowing is more prominent on the first assessment in those CIND patients, the implementation of processng speed measures in clinical neuropsychological practice would be validated. Changes in the contributions that are made by the three components of processing speed (i.e., motor speed, processing rate, and processing capacity) could also be examined. For example, are changes in processing capacity across time more pronounced than changes in processing rate or motor speed in CIND patients? My plan is to continue recruiting CIND patients in order to pursue these questions in the future. Generalized Slowing in 75 REFERENCES Adams, I. (1987). Plasticity of the synaptic contact zone following loss of synapses in the cerebral cortex of aging humans. Brain Research, 427, 343-351. Albert, M . (1993). Neuropsychological and neurophysiological changes in healthy adult humans across the age range. Neurobiology of Aging, 14, 623-625. Anastasi, A. (1988). Psychological testing (6th ed.). New York: Macmillan Publishing Company. American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, D.C.: American Psychiatric Association. Baddeley, A. D., & Hitch, G.J. (1973). Working memory. In G. A. Bower (Ed.), The psychology of learning and motivation (Vol. 8, pp. 47-90). New York: Academic Press. Baddeley, A- D. (1986). Working memory. New York: Oxford University Press. Baltes, P. B., Dittmann-Kohli, F., & Dixon, R.A. (1984). New perspectives on the development of intelligence in adulthood: Toward a dual process conception and a model of selective optimization with compensation. In P. B. B. O. G. Brim (Ed.), Life-span development and behavior (Vol. 6, pp. 33-76). New York: Academic Press. Birren, J. E. (1974). Translations in gerontology - from the lab to life: Psychophysiology and speed of response. American Psychologist, 29, 808-815. Blackford, R. C , & La Rue, A. (1989). Criteria for diagnosing age-associated memory impairment. Developmental Neuropsychology, 5, 295-306. Bors, D. A., & Forrin, B. (1995). Age, speed of information processing, recall, and fluid intelligence. Intelligence, 20, 229-248. Bowen, J., Terri, L., Kukull, W., McCormick, W., McCurry, S.M., & Larson, E.B. (1997). Progression to dementia in patients. Lancet, 349, 763-765. Brink, T. L., Yesavage, J.A., Lum, O., Heersema, P.H., Adey, M. , & Rose, T.S, (1982). Screening tests for geriatric depression. Clinical Gerontologist, 1, 37-43. Caine, E. D., Grossman, H., & Lyness, J.M. (1995). Delerium, dementia, and amnestic and other cognitive disorders and mental disorders due to a general medical condition. In B. J. H.I. Kaplan, Sadock (Ed.), Comprehensive textbook of psychiatry VI (pp. 705-754). Baltimore, M.D.: Williams and Wilkins. Generalized Slowing in 76 Cerrella, J., Poon, L.W., & Wiiliams, D.M. (1980). Age and the complexity hypothesis. In L. W. Poon (Ed.), Aging in the 1980s: Psychological issues (pp. 332-340). Washington, D.C.: American Psychological Association. Cerrella, J., Poon, L.W., & Fozard, J.L. (1981). Mental rotation and age reconsidered. Journal of Gerontology, 36, 620-624. Cohen, J., & Cohen. P. (1983). Applied multiple regression/correlation analysis for the behavioral sciences. (2nd ed.). Hillsdale, N.J.: Lawrence Erlbaum Associates, Publishers. Collins, L. F., & Long, C.J. (1996). Visual reaction time and its relationship to neuropsychological test performance. Archives of Clinical Neuropsychology, 11(7), 613-623. Corwin, J., & Bylsma, F.W. (1993). Translations of excerpts from Andre Rey's psychological examination of traumatic encephalopathy and P.A. Osterrieth's the Complex Figure Copy Test. The Clinical Neuropsychologist, 7, 3-15. Cragg, B. G. (1975). The density of synapses and neurons in normal, mentally defective and ageing human brains. Brain, 98, 81-90. Craik, F. I. M. , & Jennings, J.M. (1992). Human memory. In F. I. M . C. T. A. Salthouse (Ed.), The handbook of aging and cognition (pp. 51-110). Hillsdale, New Jersey: Lawrence Erlbaum Associates, Publishers. Crook, T., Bartus, R.T., Ferris, S.H., Whitehouse, P., Cohen, G.D., & Gershon, S. (1986). Age-associated memory impairment: Proposed diagnostic criteria and measures of clinical change. Report of a National Institute of Mental Health Work Group. Developmental Neuropsychology, 2, 261-276. DeKosky, S. T. P., A . M . (1994). Neurochemisty of aging. In M . L. A. J. E. Knoefel (Ed.), Clinical neurology of aging (2nd ed., pp. 79-101). New York: Oxford University Press. Delis, D. C , Kramer, J.H., Kaplan, E., & Ober, B.A. (1987). California Verbal Learning Test: Adult Version. San Antonio, TX: The Psychological Corporation. Delis, D. C , Massman, P.J., Butters, N. , Salmon, D.P., Cermak, L.S., & Kramer, J.H. (1991). Profiles of demented and amnesic patients on the California Verbal Learning TestL Implications for the assessment of memory disorders. Psychological Assessment, 3(1), 19-26. Devand, D. P., Folz, M. , Gorlyn, M. , Moeller, J.R., & Stern, Y. (1997). Questionable dementia: Clinical course and predictors of outcome. Journal of the American Geriatric Society, 45, 321-328. Generalized Slowing in 77 Feldman, H., Schulzer, M. , Wang, S., Tuokko, FL, & Beattie, B.L. (Ed.). (1995). The Functional Rating Scale in Alzheimer's disease assessment: A longitudinal study: John Wiley & Sons Ltd. Feldman, H., Beattie, B.L., Hayden, S., Sadovnik, A.D., Studney, D., & Beach, T. (1996). The consortium of Canadian centres for clinical cognitive research: A cohort study of cognitive impairment and dementia syndromes with the development of a consortium registry. Vancouver, B.C.: University of British Columbia. Ferris, S., Crook, T., Sathanathan, G., & Gershon, S. (1976). Reaction time as a diagnostic measure in senility. Journal of the American Geriatrics Society, 24(12), 529-533. Ferris, S. H. K., A. (1996). Commentary on age-associated memory impairment, age-related cognitive decline and mild cognitive impairment. Aging, Neuropsychology, and Cognition, 3(2), 148-153. Flicker, C , Ferris, S.H., & Reisberg, B. (1993). A longitudinal study of cognitive function in elderly persons with subjective memory complaints. Journals of the American Geriatric Society, 41, 1029-1032. Folstein, M . F., Folstein, S.E., & McHugh, P.R. (1975). Mini-mental state. Journal of Psychiatric Research, 12, 189-198. Graf, P., & Schacter, D.L. (1985). Implicit and explicit memory for new associationsin normal and amnesic subjects. Journal of Experimental Psychology: Learning, Memory, and Cognition, 11, 501-518. Graf, P., & Uttl, B. (1995). Component processes of memory: Changes across the adult life-span. Swiss Journal of Psychology, 54(2), 113-130. Graf, P., Squire, L.R., & Mandler, G. (1984). The information that amnesic patients do not forget. Journal of Experimental Psychology: Learning, Memory, and Cognition, 10, 164-178. Graf, P., Shimamura, A.P., & Squire, L.R. (1985). Priming across modalities and priming across category levels: extending the domain of preserved function in amnesia. Journal of Experimental Psychology: Learning, Memory, and Cognition, 11, 385-295. Graham, J. E., Rockwood, K., Beattie, B.L., Eastwood, R , Gauthier, S., Tuokko, H., & McDowell, I. (1997). Prevalence and severity of cognitive impairment with and without dementia in an elderly population. Lancet, 349,1793-1796. Hebb, D. O. (1942). The effect of early and late brain injury upon test scores and the nature of normal adult intelligence. Proceedings of the American Philosophical Society, 85, 275-292. Generalized Slowing in 78 Horn, J. L., & Cattell, R.B. (1963). Age differences in fluid and crystallized intelligence. Acta Psychologica, 26, 107-129. Howieson, D. B., Holm, M.A., Kaye, J.A., Oken, B.S., & Howieson, J. (1993). Neurologic function in the optimally healthy oldest old: Neuropsychological evaluation. Neurology, 43,1882-1886. Hughes, C. P., Berg, L., & Danzinger, W.L. (1982). A new clinical scale for the staging of dementia. The British Journal of Psychiatry, 140, 566-72. Huppert, F. A., Brayne, C , & O'Connor, D.W. (1994). Dementia and normal aging. Cambridge: Cambridge University Press. Huttenlocher, P. R. (1979). Synaptic density in human frontal cortex-developmental changes and effects of aging. Brain Research, 163,195-205. Huynh, H., & Feldt, L.S. (1976). Estimation of the Box correction for degrees of freedom in randomized block and split-plot designs. Journal of Educational Statistics, 1, 29-82. Ivnik, R. J., Malec, J.F., Smith, G.E., Tangalos, E.G., Petersen, R.C., Kokmen, E., & Kurland, L.T. (1992). Mayo's older Americans normative studies: WAIS-R, WMS-R and AVLT norms for ages 56 through 97. The Clinical Neuropsychologist (Suppl), 6,1-104. Jacobs, D. M. , Sano, M. , Dooneief, G., Marder, K., Bell, K.L., & Stern, Y. (1995). Neuropsychological detection and characterization of preclinical Alzheimer's disease. Neurology, 45, 957-962. Johnson, S. H. R., J.M. (Ed.). (1993). A cognitive neuroscience perspective on age-related slowing: Developmental changes in the functional architecture. San Diego: Academic Press Inc. Larrabee, G., Levin, H.S., & High, W.M. (1986). Senescent forgetfulness: A quantitative study. Developmental Neuropsychology, 2, 373-385. Lezak, M . D. (1995). Neuropsychological Assessment. (3rd ed.). New York: Oxford University Press. Lindberger, U., Mayr, U., & Kliegl, R. (1993). Speed and intelligence in old age. Psychology and Aging, 8,207-220. Masliah, E., Mallory, M. , Hansen, L., DeTeresa, R., & Terry, R.D. (1993). Quantitative synaptic alterations in the human neocortex during normal aging. Neurology, 43, 192-197. Generalized Slowing in 79 McKhann, G., Drachman, D., Folstein, M. , Katzman, R., Price, D., Stadlan, E.M. (1984). Clinical diagnosis of Alzheimer's disease: Report of the NINCDS- ADRD A Work Group under the auspices of the Department of Health and Human Services Task Force on Alzheimer's Disease. Neurology, 34, 939-944. Meyers, J., & Meyers, K. (1995). The Rey-Osterrieth Complex Figure Test and Recognition Trial: Professional manual. Odessa, FL: Psychological Assessment Resources, Inc. Miller, A. K. H., Alston, R.L., & Corsellis, J.A.N. (1980). Variation with age in the volume of grey and white matter in the cerebral hemispheres of man: Measurements with an image analyser. Neuropathology and Applied Neurobiology, 6, 119-132. Mitrushina, M . N. , Boone, K.B., D'Elia, L.F. (1999). Handbook of normative data for neuropsychological assessment. New York: Oxford University Press. Morris, J. C , McKeel, D.W., Storandt, M. , Rubin, E.H., Price, J.L., Grant, E.A., Ball, J.M., & Berg, L. (1991). Very mild Alzherimer's disease: Informant-based clinical, psychometric, and pathological distinction from normal aging. Neurology, 41, 469-478. Muller, G., Richter, R.A., Weisbrod, S., & Klingberg, F. (1991). Reaction time prolongation in the early stage of presenile onset Alzheimer's disease. European Archives of Psychiatry and Clinical Neuroscience, 241,46-48. Nebes, R. D., Madden, D.J. (1988). Different patterns of cognitive slowing produced by Alzheimer's disease and normal aging. Psychology and Aging, 3(1), 102-104. Nebes, R. D., Brady, C.B., & Reynolds, C.F. (1992a). Cognitive slowing in Alzheimer's disease and geriatric depression. Journal of Gerontology: Psychological Sciences. 47(5), P331-P336. Nebes, R. D., & Brady, C.B. (1992b). Generalized cognitive slowing and severity of dementia in Alzheimer's disease: Implications for the interpretation of response-time data. Journal of Clinical and Experimental Neuropsychology, 14(2), 317-326. Paolo, A. M. , Troster, A.I., & Ryan, J.J,. (1997). California Verbal Learning Test: Normative data for the elderly. Journal of Clinical and Experimental Neuropsychology, 19(2), 220-234. Park, D. C , Lautenschlager, Smith, A.D., Earles, J.L., Frieske, D., Zwahr, M. , & Gaines, C L . (1996). Mediators of long-term memory performance across the life span. Psychology and Aging, 11(4), 621-637. Generalized Slowing in 80 Pate, D. S., & Margolin, D.I. (1994a). Cognitive slowing in Parinson's and Alzheimer's patients: Distinguishing bradyphrenia from dementia. Neurology, 44, 669-674. Pate, D. S., Margolin, D.I., Friedrich, F.J., & Bentley, E.E. (1994b). Decision-making and attentional resources in ageing and in dementia of the Alzheimer's type. Cognitive Neuropsychology, 11(3), 321-339. Pirozzolo, F. J., Christensen, K.J., Ogle, K.M. , Hansen, E.C., & Thompson, W.G. (1981). Simple and choice reaction time in dementia: Clinical implications. Neurobiology of Aging, 2. 113-117. Rediess, S., & Caine, E.D. (1996). Aging, cognition, and DSM-IV. Aging, Neuropsychology, and Cognition, 3(2), 105-117. Reisberg, B., Ferris, S.H., De Leon, M.J., Crook, T. (1982). The Global Deterioration Scale for assessment of primary degenerative dementia. American Journal of Psychiatry, 139(9), 1136-1139. Salthouse, T. A. (1980). Age and memory: Strategies for localizing the loss. In J. L. F. L.W. Poon, L. Cermak, D. Arenberg, & L.W. Thompson (Ed.), New directions in memory and aging (pp. 47-65). Hillsdale N.J.: Erlbaum. Salthouse, T. A. (1985). A theory of cognitive aging. (Vol. 28). Amsterdam: North Holland. Salthouse, T. A., & Babcock, R.L. (1991). Decomposing adult age differences in working memory. Developmental Psychology, 27(5), 763-776. Salthouse, T. A. (1991a). Theoretical perspectives on cognitive aging. Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers. Salthouse, T. A. (1991b). Mediation of adult age differences in cognition by reductions in working memory and speed of processing. Psychological Science, 2(3), 179-183. Salthouse, T. A. (1992a). Mechanims of age-cognition relations in adulthood. Hillsdale, New Jersey: Lawrence Erlbaum Associates, Publishers. Salthouse, T. A. (1992b). Influence of processing speed on adult age differences in working memory. Acta Psychologica, 79, 155-170. Salthouse, T. A. (1993). Speed mediation of adult age differences in cognition. Developmental Psychology. 29(4), 722-738. Generalized Slowing in 81 Salthouse, T. A. (1994a). Aging associations: Influence of speed on adult age differences in associative learning. Journal of Experimental Psychology: Learning, Memory, and Cognition, 20(6), 1486-1503. Salthouse, T. A. (1994b). The nature of the influence of speed on adult age differences in cognition. Developmental Psychology, 30(2), 24-259. Salthouse, T. A. (1996a). The processing speed theory of adult age differences in cognition. Psychological Review, 103(3), 403-428. Sliwinski, M. , & Buschke, H. (1996). Processing speed and memory in aging and dementia. Journal of Gerontology: Psychological Sciences, 52B(6), P308-P318. Smith, G. E., Peterson, R.C., Parisi, J.E., Ivnik, R.J., Kokmen, E., Tangalos, E.G., & Waring, S. (1996). Definition, course, and outcome of mild cognitive impairment. Aging, Neuropsychology, and Cognition, 3(2), 141-147. Spreen, O., & Strauss, E. (1998). A compendium of neuropsychological tests: Administration, norms, and commentary. New York: Oxford University Press. Tierney, M . C , Szalai, J.P., & Snow, W.G. (1996). Prediction of probable Alzheimer's disease in memory-impaired patients: A prospective longitudinal study. Neurology, 46, 661-665. Tuokko, H., & Crockett, D. (1989). Cued recall and memory disorders in dementia. Journal of Clinical and Experimental Neuropsychology, 11(2), 278-294. Tuokko, H., Vernon-Wilkinson, R., Weir, J., & Beattie, B.L. (1991). Cued recall and early identification of dementia. Journal of Clinical and Experimental Neuropsychology, 13(6), 871-879. Verhaeghen, P., & Salthouse, T.A. (1997). Meta-analyses of age-cognition relations in adulthood: Estimates of linear and nonlinear age effects and structural models. Psychological Bulletin, 122(3), 231-249. Vernon-Wilkinson, R., & Tuokko, H. (1990, October). Factors affecting test scores in neuropsychological assessment. Monthly Newsletter of the College of Psychotherapists of British Columbia, 11-13. Wechsler, D. (1981). Wechsler Adult Intelligent Scale-Revised manual. New York: The Psychological Corporation. Williams, P. A., Jones. G.H., Briscoe, M. , Thomas, R., & Cronin, P. (1991). P300 and reaction-time measures in senile dementia of the Alzheimer type. British Journal of Psychiatry. 159,410-414. Generalized Slowing in 82 Yesavage, J. A., Brink, T.L., Rose, T.L., & Adey, M . (1986). The geriatric depression rating scale: Comparison with other self-report and psychiatric rating scales. In L. W. Poon (Ed.), Handbook of Clinical Memory Assessment of Older Adults (pp. 153-167). Washington, D.C.: American Psychological Association. Generalized Slowing in APPENDIX A The two Clinic for Alzheimer's Disease and Related Disorders informed consent forms that all participants are required to fill out prior to their clinic assessment. VANCOUVER HOSPITAL AND HEALTH SCIENCES CENTRE UBC SITE ALZHEIMER CLINIC AUTHORIZATION FOR RELEASE OF INFORMATION ON THE HEALTH RECORD Authorization is hereby granted to release information from my Health Record to the following: Dr. B. Lynn Beattie Clinical Director Alzheimer Clinic Vancouver Hospital & Health Sciences Centre 2211 Wesbrook Mall Vancouver, B.C. V6T 2B5 Witness: Date: Signed: Relationship to patient: B:\RELINFO.FRM CLINIC FOR ALZHEIMER'S DISEASE f~>***s-^' \ . AND RELATED DISORDERS VANfOUVPP HOSPTTA] G - 3 5 PURDY PAVILION m n w u f t i v n u j n i A L A U U B I N G U O S P H U I K U U I E D W I T H TEL: (604) 8 2 2 - 7 0 3 1 & H e a l t h S c i e n c e s m m m m 0 f , m H , { ( [ | M | U F A X : ( 6 0 4 ) 8 2 2 - 7 1 9 1 CONSENT FORM: All patients will be asked to sign this consent form at the time of their assessment and a copy will be provided to them. PRINCIPAL INVESTIGATOR: Dr. Howard Feldman CO-INVESTIGATORS: Dr. B.L. Beattie, Dr. S. Hayden, Dr. M. Genge, Dr. D. Foti Dr. D. Sadovnick TELEPHONE: 822-7031 TITLE: Clinical Studies in Cognitive impairment and Dementia Syndromes In order to gather as much information as possible about patients who have been referred to the UBC Clinic for Alzheimer's Disease and Related Disorders and their families, we ask everyone who attends the clinic to cooperate with us in the gathering of basic diagnostic and demographic information. This information will be collected in an anonymous data base. Strict confidentiality is applied to all gathered information. Personal identification of patients and family members is available only to members of the UBC Alzheimer Clinic team. Whenever information is shared with collaborators outside of this immediate group, the information is coded to respect the personal identity of the patient and family to ensure confidentiality. RECONTACT The Alzheimer's clinic staff may re-contact you, your caregiver, or your next of kin to update your condition. You may be made aware of research projects that are being conducted in the clinic. Each specific project will have its own description and you may be asked to participate. A separate consent form would be signed for such projects. You may decline to enter into or may withdraw from this or any of these projects at any time without any consequence to continuing medical care and/or possible eligibility to participate in future projects. Clinical Studies in Cognitive Impairment and Dementia Syndromes- Page 1 of 2 May 1997 • 2 2 H W [ $ I « O O I - M m V A N C O U V E R , B S M I S H C O I U H H U V 6 T 2 B 5 T E L : ( 6 0 A ) 822 -71 21 F A X : ( 60 t ) 8 2 2 - 7 2 6 8 POTENTIAL BENEFITS AND RISKS: Information learned from clinic patients and their families will be useful in the understanding, diagnosis and treatment of cognitive impairment (problems with memory, thinking and behavior) and dementia. The gathering of this information will not expose you to any risk. Additional research projects will be fully explained to you, including the potential benefits and risks of that project. If you would like more detail about something mentioned here, or need information not included here, you should feel free to ask the investigators and clinic staff. If you have any questions regarding your rights as a participant in the study, you may also contact Dr. Richard Spratlev, UBC Office of Research Services at 822-8584. I, . • • agree to allow the UBC Clinic for Alzheimer's Disease and Related Disorders to use information collected during my visit for the purposes of research into cognitive impairment and dementia syndromes. Participant Signature: . Print Name: • Date: -Caregiver/Next of Kin (if required):_ • • -Print Name: : Date: Witness Signature: Print Name: Date: Contact persons for patients are: cc: Participant Alzheimer Clinic Chart Family Physician Dr. Howard Feldman, MD, 822-7697 Dr. B.L. Beattie, MD, 822-7031 Gilda Sam, RN, 822-7176 Clinical Studies in Cognitive Impairment and Dementia Syndromes- Page 2 of 2 May 1997 Generalized Slowing in 87 APPENDIX B Clinic for Alzheimer's Disease and Related Disorders, University of British Columbia Hospital Clinical Neuropsychological Test Battery Geriatric Depression Scale* WAIS-R Information Subtest* WAIS-R Digit Span Subtest* Rey-Osterrieth Complex Figure Copy Immediate Recall WMS-R Logical Memory Test Immediate Recall WAIS-R Similarities Subtest* WAIS-R Comprehension Subtest* Select enough of the following to complete a 30 min delay for the Rey-Osterrieth Figure and Logical Memory WAIS-R Vocabulary and/or WAIS-R Arithmetic Subtest WAIS-R Picture Completion and/or WAIS-R Picture Arrangement Rey-Osterrieth Complex Figure: Delayed Recall (30 min) WMS-R Logical Memory Test: Delayed Recall (30 min) WAIS-R Object Assembly Trail Making Test A and B Buschke Cued Recall Test* or California Verbal Learning Test Finger Tapping Test* Immediate Recall WAIS-R Block Design* WAIS-R Digit Symbol Substitution Subtest* Buschke Cued Recall Test (OVER) or California Verbal Learning Test: Generalized Slowing in 88 Delayed Recall* Delayed Recall (20 min) If additional test is necessary for design memory, do LNNB Memory for Designs or WMS-R Visual Reproduction Immediate Recall Controlled Oral Word Association (FA.S.) Test* LNNB Memory for Designs Delayed Recall (5 min) Stroop Color Word Test WMS-R Visual Reproduction Delayed Recall (30 min) Benton Visual Retention Test: Multiple Choice Form* Select from the following as appropriate: Wisconsin Card Sorting Test or Booklet Category Test Raven's Colored Progressive Matrices or Hooper Visual Organization Test Note: Test instruments that were used in this research are indicated in bold. WAIS-R := Wechsler Adult Intelligence Scale-Revised, WMS-R = Wechsler Memory Scale-Revised, LNNB = Luria-Nebraska Neuropsychological Battery * Indicates the Standard Test Battery, all other tests are part of the Extended Test Battery 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0088987/manifest

Comment

Related Items