DOES IMPAIRED EXECUTIVE FUNCTIONING DIFFERENTIALLY IMPACT VERBAL MEMORY MEASURES IN OLDER ADULTS WITH SUSPECTED DEMENTIA?

Transcription

1 The Clinical Neuropsychologist, 20: , 2006 Copyright # Taylor and Francis Group, LLC ISSN: print= online DOI: / DOES IMPAIRED EXECUTIVE FUNCTIONING DIFFERENTIALLY IMPACT VERBAL MEMORY MEASURES IN OLDER ADULTS WITH SUSPECTED DEMENTIA? Brian L. Brooks 1, Linda E. Weaver 1,2, and Charles T. Scialfa 1 1 University of Calgary, Department of Psychology, Calgary, AB, Canada, and 2 Rockyview General Hospital, Seniors Health Ambulatory Care, Calgary, AB, Canada The purpose of this study was to examine whether executive dysfunction differentially impacts list-learning and story recall tasks in a sample of older adults referred for suspected cognitive impairment. Older adults (N ¼ 61) with mild cognitive impairment (MCI) or probable mild dementia, and those who did not meet criteria for diagnosis of dementia, were assessed using measures of executive function and verbal memory. Two groups were established based on performance on measures of executive function: (a) the No Executive Dysfunction group (NoED; n ¼ 33) consisted of persons without impairment on any obtained measures of executive function; and (b) the Executive Dysfunction group (ED; n ¼ 28) contained persons with impairment on at least one of the measures of executive function. The two groups were compared on performance on two measures of verbal memory, the California Verbal Learning Test II (CVLT-II) and the Logical Memory (LM) subtest from the Wechsler Memory Scale Revised (WMS-R). The NoED group performed significantly better than the ED group on the total learning and short delay free recall trials of the CVLT-II. However, there were no significant differences between the groups on the other indices of the CVLT-II (i.e., long delay free recall, recognition, recall repetitions, recall intrusions, or recognition false-positives) or on the immediate and delayed recall trials of the LM measure. These results support previous research demonstrating the impact of executive dysfunction on the acquisition of and short-delay retrieval of verbal information in older adults with suspected cognitive impairment. INTRODUCTION Deficits on measures of memory are a criterion for the diagnosis of probable dementia across various dementia etiologies (American Psychiatric Association; APA, 1994). Memory impairment is also an important criterion for determining the presence of MCI (Petersen et al., 2001). Memory impairments in older adults with MCI or probable dementia are observed in their performance on measures in neuropsychological assessment, with formal verbal memory measures including the California Verbal Learning Test (CVLT; Delis, Kramer, Kaplan, & Ober, 1987, 2000) and the Wechsler Memory Scale (WMS; Wechsler, 1987, 1997) and its subtests. Address correspondence to: Brian L. Brooks, University of Calgary, Department of Psychology, 2500 University Drive NW, T2N 1N4 Calgary, Canada. Tel.: Fax: blbrooks@ucalgary.ca Accepted for publication: October 5,

2 EXECUTIVE DYSFUNCTION AND MEMORY 231 Although both the CVLT and WMS are tests of verbal memory initially thought to be interchangeable (e.g., Delis, Cullum, Butters, Cairns, & Prifitera, 1988), they are not always consistent in their results in clinical populations. For example, Randolph et al. (1994) identified discrepancies in performance between different measures of verbal memory in a control and patient sample consisting of those with temporal-lobe epilepsy and schizophrenia. It may be that performance on the tests can be differentially affected by the co-existence of cognitive impairment, an issue that is particularly relevant to dementia. For example, deficits in memory and executive function, which refers to those higher cognitive capacities that allow a person to initiate, monitor, and regulate their own activity in order to engage in independent and purposive behavior (Lezak, 1995), have been identified as important predictors of the progression from MCI to probable dementia (Albert, Moss, Tanzi, & Jones, 2001). Recent studies have examined whether deficits on measures of executive function have detrimental impact on performance on verbal memory measures in various clinical populations. Vanderploeg, Schinka, and Retzlaff (1994) examined performance on the CVLT and its relationship to executive functioning in a sample of 115 people with traumatic brain injury or various neurological diseases. The authors obtained a five-factor structure of the CVLT, similar to that reported by Delis et al. (1988), which provided two factors (verbal learning and working memory) that were significantly correlated with measures of executive functions. In persons with Parkinson s disease, Ringe, Frol, Saine, and Cullum (1999) found that measures of executive function were positively correlated with short delay recall scores on both the CVLT and a measure of visual memory. In a regression analysis, executive functions accounted for 18% of the variance in the memory measures. Fossati, Coyette, Ergis, and Allilaire (2002) measured executive functions and episodic verbal memory performance in a life-span adult sample of inpatients with diagnoses of unipolar or bipolar depression. On a verbal memory measure involving free and cued recall of words (i.e., a verbal memory task adapted by the authors), those depressed persons with impaired executive functions displayed poorer performance across the indices than those persons without impaired executive functions. It has also become important to examine whether executive dysfunction negatively impacts different measures of verbal memory in a similar manner. Tremont, Halpert, Javorsky, and Stern (2000) found that executive dysfunction differentially impacted performance on the CVLT and LM. Patients with various medical (largely neurological and psychiatric) disorders were classified into groups as either minimal executive dysfunction (MED) or significant executive dysfunction (SED) based on their performance on measures of executive functioning. Tremont et al. reported that the SED group performed significantly worse than the MED group on the total words learned and most recall conditions of the CVLT; however, there was no difference between the groups on the LM test. This study by Tremont et al. demonstrated that executive dysfunction, defined as performance below age- and education-expected norms on executive function measures, could differentially impact verbal memory measures. Tremont et al. provide some insight into the potential neuropsychological causes of discrepancy (i.e., executive dysfunction) between performances on the CVLT compared to the LM in a clinical population. It may be that the story-recall format provides contextual cues that aid in the learning

3 232 BRIAN L. BROOKS ET AL. and recall of information, whereas list learning does not contain the same cues and one must rely more heavily on their own abilities to organize the material for optimal learning and recall. The relationship between memory performance and executive functions has yet to be investigated in older adults with functional decline or neurological impairments. Given that measures of verbal memory play a critical role in the assessment of older persons with suspected impairments, this study undertook to examine the differential impact of executive dysfunction on the performance of measures of verbal memory in older persons referred for neuropsychological assessment for progressive dementia and other neurological disorders. It was hypothesized that the presence of executive dysfunction would negatively impact performance on the CVLT-II measure, but not the LM measure, in this population of older adults. METHODS Participants Data for this research was obtained from archival neuropsychological assessments (n ¼ 85) contained in the Seniors Health program at the Rockyview General Hospital in Calgary, Alberta, Canada. Participants were referred to the program by family physicians for the investigation of suspected memory decline or general cognitive impairment. Neuropsychological charts were included in the study if the participants were between the ages of 55 and 90 years and were able to complete the battery of neuropsychological tests administered. From a pool of referrals to a neuropsychology service, participants ranged from patient controls (n ¼ 15; those cases for whom the physician, patient, or family were concerned but criteria for diagnosis of MCI or dementia were not met), those with MCI (n ¼ 20; those cases for whom there was an objective decline in cognitive functioning, predominantly observed as memory impairment and may include other domains, but did not have social or occupational impairments to warrant a diagnosis of dementia; Petersen et al., 1999, 2001), to those with a diagnosis of probable mild dementia (n ¼ 26; those cases for whom dementia of varied aetiologies was diagnosed using criteria from the Diagnostic and Statistical Manual of Mental Disorders, fourth edition; APA, 1994). Consistent with many clinical settings, diagnoses were given by a neuropsychologist based on charted medical history, informant and patient report, and formal assessment including neuropsychological tests. Etiology of dementia was based on neuropsychological performance, medical history, and available neuroimaging findings. For the probable dementia group, etiologies included Alzheimer s disease (50%), vascular dementia (27%), and mixed Alzheimer s þ vascular dementia (23%). Participants were excluded if there were incomplete data available for the measures of executive functions and=or or verbal memory (n ¼ 16) or if they presented with global cognitive impairment across all measures of executive functions (n ¼ 8) in order to avoid floor effects. Neuropsychological Measures Measures selected for this study were part of a larger geriatric neuropsychological battery comprised of standardized neuropsychological measures and were

4 EXECUTIVE DYSFUNCTION AND MEMORY 233 chosen to assess executive functions and memory performance. Age-equivalent standard scores for all tests were calculated from the raw scores using normative data to account for any age- and=or education-related differences in performance, as well as to make the presentation of the data more accessible for clinicians. For the purpose of easier cross-measure comparisons, all normative scores were converted into T-scores. Tests selected to evaluate executive function (Lezak, 1995) included: (a) Wisconsin Card Sorting Test-64 (WCST-64) Perseverative Responses (Kongs, Thompson, Iverson, & Heaton, 2000) as a measure of cognitive flexibility and abstract problem-solving; (b) Trail Making Test Version B (Trails B) standard score for total time (Ivnik, Malec, Smith, Tangolos, & Petersen, 1996; Reitan & Wolfson, 1985) as a measure of set shifting and mental tracking; and (c) Controlled Oral Word Association Test (COWAT) standard score for total words produced beginning with the letters F, A, and S (Benton & Hamsher, 1976; Gladsjo et al., 1999) as a measure of verbal fluency and set shifting. It should be noted that Trails B was discontinued after 360 seconds. In addition, this was not an exhaustive list of measures designed to assess impairments in executive functions, but rather reflects measures commonly employed in this outpatient clinic. Tests selected to measure verbal memory included: (a) California Verbal Learning Test II (CVLT-II; Delis et al., 2000) standard scores for total words recalled across the five learning trials, short delay free recall after a second list is presented once, long delay free recall after a 20 minute delay, yes=no recognition of words, total recall repetition errors, total recall intrusion errors, and recognition false positives; and (b) Logical Memory (LM) standard scores for immediate and delayed (30-minute delay) recall subtests from the Wechsler Memory Scale Revised (WMS-R; Wechsler, 1987) with more recent normative data for the LM subtest from the Mayo s Older Americans Normative Studies (MOANS; Ivnik et al., 1992). Groups Impairment on executive function measures, defined as performance below the 10th percentile compared to the standardized peer group, was used to create the comparison groups used here. Based on performance on the selected executive function measures, two groups were established; (a) a no executive dysfunction (NoED) group (n ¼ 33), which included those persons who did not have impaired performance on any of the selected executive functioning measures; and (b) an executive dysfunction (ED) group (n ¼ 28), which included those persons who had impairment on either one or two of the executive functioning measures. The decision to group persons with impairment on either one or two executive function measures into one ED group was based on preliminary analyses, which indicated that these two groups did not differ significantly across any of the executive function or memory measures. Persons with impairment across all three measures of executive function were not included in this retrospective study, as the sample size was small (n ¼ 8), they tended to have more severe global impairment, and we wanted to avoid floor effects on the measures (Fox, Olin, Erblich, Ippen, & Scheider, 1998).

5 234 BRIAN L. BROOKS ET AL. RESULTS Univariate t-tests were used to examine group differences on demographic information. Multivariate t tests (Hotelling s T 2 ) were used to examine whether the groups differed on the measures. Hotelling s T 2 were selected in order to protect Type I error rate as well as to account for the correlations between the different scores within each measure of memory (Weinfurt, 1995). Three separate multivariate analyses were conducted to test for group differences on the executive and memory measures; therefore, to further protect Type I error rate, significance for the multivariate t-tests was set at p <.05=3 ¼.017. Since it was hypothesized that the ED group would perform significantly worse than the NoED group on the memory measures, p values for the t-tests represent one-tailed level of significance. Nonparametric analyses were conducted using chi-square analyses. Demographics Participant demographics are presented in Table 1. T-tests revealed nonsignificant univariate effects for age (t [59] ¼ 0.264, p ¼.793) and education (t [59] ¼ 1.433, p ¼.157). Chi-square analyses for gender revealed there were no significant differences between women and men in the NoED group (v 2 [1] ¼ 1.485, p ¼.223); however, there were significantly more women than men in the ED group (v 2 [1] ¼ 5.143, p ¼.023). Regarding diagnoses, there were no significant differences Table 1 Participant demographics for the groups NoED group ED group n Mean age (yrs) SD Range Education (yrs) SD Range Gender (%): Male Female Hand (%): Right Left Both Unknown Diagnoses (%): Patient Control Mild Cognitive Impairment Mild Dementia Note. NoED Group refers to those participants without any impairment on measures of executive functions, while the ED Group had one or two scores below the 10th percentile on measures of executive functions.

6 EXECUTIVE DYSFUNCTION AND MEMORY 235 in diagnoses for the NoED group (v 2 [2] ¼.545, p ¼.761), but there were significantly more participants with mild dementia than patient controls in the ED group (v 2 [2] ¼ 7.357, p ¼.025). As a result of the differences in number of participants with mild dementia diagnoses in the ED group, the possibility of general severity, rather than executive dysfunction, impacting verbal memory performance was explored. In order to examine this issue, measures that are known to decline with general severity associated with dementia were compared across the two groups. For example, on a measure of language abilities (Boston Naming Test; Kaplan, Goodglass, & Weintraub, 1978), there was no significant difference between the NoED (mean ¼ 8.53, SD ¼ 3.83) and ED (mean ¼ 8.92, SD ¼ 3.60) groups (F [1, 55] ¼.152, p ¼.70). As well, on a measure of visual-spatial abilities (e.g., Line Bisection), again there was no significant difference between the NoED (mean ¼ 3.57, SD ¼ 5.02) and ED (mean ¼ 3.38, SD ¼ 2.35) groups (F [1, 47] ¼.025, p ¼.87). As a result of the absence of differences on these measures, it is less likely that any differences obtained on measures of verbal memory are attributable to general severity of cognitive decline. Table 2 presents the raw and standardized T-scores for all of the executive function and verbal memory measures for the NoED and ED groups. Statistical analyses comparing the groups used the normative scores for the CVLT-II and LM in order to increase the palatability of results for clinicians. However, it should be noted that analyses with the raw data mirror the results using normative scores. Figure 1 presents mean standard scores for the two groups on the CVLT-II Total Learning, Short Delay Free Recall, Long Delay Free Recall and Recognition. Hotelling s T 2 revealed that the ED group differed significantly from the NoED group (t [56] ¼ 2.870, p ¼.016). Follow-up t tests revealed that the ED group performed significantly lower than the NoED group on the CVLT-II Total Learning (t [59] ¼ 3.335, p ¼.001, w 2 ¼.167) and CVLT-II Short Delay Free recall (t [59] ¼ 2.534, p ¼.007, w 2 ¼.097); however, the groups were not significantly different on the CVLT-II Long Delay Free Recall (t [59] ¼ 1.794, p ¼.039, w 2 ¼.047) or Recognition (t [59] ¼ 1.635, p ¼.054, w 2 ¼.048) trials. Hotelling s T 2 also revealed there were no significant group differences (t [57] ¼ 0.754, p ¼.263, w 2 ¼.046) on the CVLT-II Total Recall Repetitions, Total Recall Intrusions, and Recognition False Positives. Figure 2 presents the Logical Memory Immediate and Delayed Recall T-scores for the groups. Hotelling s T 2 revealed that the groups were not significantly different from each other on either the Immediate or Delayed portions of the Story Recall (t [58] ¼ 0.081, p ¼.462, w 2 ¼.003). These results suggest level of executive dysfunction did not impact the ability to recall a newly learned story after an immediate or longer time delay. Correlations between the executive function and memory measures are presented in Table 3. Given that it was hypothesized that as executive function decreased, performance on memory measures would also decrease, one-tailed correlations were examined. Trails B performance was positively correlated with CVLT-II Total Learning and LDFR scores; COWAT scores were positively correlated with the CVLT SDFR, LDFR, Repetition, and Intrusion scores; and WCST Perseverative Responses were not significantly related to any memory scores. There were no significant correlations between the executive function measures and the LM measures.

7 236 BRIAN L. BROOKS ET AL. Table 2 Mean raw and T-scores (standard deviation) on measures of executive functions and verbal memory for the two groups NoED group ED group Raw score T-Score Raw score T-Score Trails B (43.51) a (7.96) (71.04) a (9.93) COWAT (10.12) b (8.13) (11.86) b (10.83) WCST-PR (12.98) c (13.28) (19.48) c (14.75) CVLT-II Total Learning (9.91) d (11.16) (12.19) d (13.12) CVLT-II SDFR 6.48 (2.63) d (15.08) 5.00 (3.19) d (11.79) CVLT-II LDFR 6.03 (3.86) d (15.31) 4.50 (3.46) d (12.90) CVLT-II Recognition (2.74) e (10.96) (3.50) e (13.10) CVLT-II Repetitions 2.62 (2.82) f (8.53) 1.69 (3.25) f (7.24) CVLT-II Intrusions 6.07 (6.34) g (16.07) 7.12 (9.12) g (16.78) CVLT-II False Positives 4.21 (3.68) h (13.00) 4.64 (3.57) h (15.97) LM Immediate Recall (11.60) i (11.88) (11.74) i (13.55) LM Delayed Recall (8.47) i (12.32) (9.07) i (11.92) Notes. NoED Group refers to those participants without any impairment on measures of executive functions, while the ED Group had one or two scores below the 10th percentile on measures of executive functions. Trails B ¼ Trail Making Test B (Reitan & Wolfson, 1985); COWAT ¼ Controlled Oral Word Association Test (Benton & Hamsher, 1976); WCST-PR ¼ Wisconsin Card Sorting Test, Perseverative Responses (Kongs et al., 2000); CVLT-II ¼ California Verbal Learning Test Second Edition (Delis et al., 2000); SDFR ¼ Short Delay Free Recall; LDFR ¼ Long Delay Free Recall; LM ¼ Logical Memory subtest (Wechsler, 1987). a Total time in seconds. b Sum of total words produced using letters F, A, and S. c Sum of Perseverative Responses. d Sum of words recalled across all trials. e Sum of words positively recognized. f Sum of repetitions across all recall trials. g Sum of intrusions across all recall trials. h Sum of false positives during recognition. i Sum of accurate parts of story recalled. Figure 1 Mean T-scores on the CVLT-II total learning, short delay free recall, long delay free recall, and recognition measures. Note. NoED group refers to those participants without any impairment on measures of executive functions, while the ED group had one or two scores below the 10th percentile on measures of executive functions; CVLT-II ¼ California Verbal Learning Test Second Edition (Delis et al., 2000).

8 EXECUTIVE DYSFUNCTION AND MEMORY 237 Figure 2 Mean T-scores on the Logical Memory immediate and delayed recall measures. Note. NoED group refers to those participants without any impairment on measures of executive functions, while the ED group had one or two scores below the 10th percentile on measures of executive functions; Logical Memory (Wechsler, 1987). Table 3 Correlations between measures of executive functions and memory Trails B COWAT WCST-PR CVLT-II Total Learning CVLT-II SDFR CVLT-II LDFR CVLT-II Recognition CVLT-II Repetitions CVLT-II Intrusions CVLT-II False Positives LM Immediate Recall LM Delayed Recall Notes. Trails B ¼ Trail Making Test B (Reitan & Wolfson, 1985); COWAT ¼ Controlled Oral Word Association Test (Benton & Hamsher, 1976); WCST-PR ¼ Wisconsin Card Sorting Test, Perseverative Responses (Kongs et al., 2000); CVLT-II ¼ California Verbal Learning Test Second Edition (Delis et al., 2000); SDFR ¼ Short Delay Free Recall; LDFR ¼ Long Delay Free Recall; LM ¼ Logical Memory subtest (Wechsler, 1987). p <.05. p <.01 for one-tailed correlations. DISCUSSION In this study, we examined the impact of executive dysfunction on performance on two verbal memory tests in a sample of older adults referred for suspected cognitive impairment. Consistent with previous studies, we found that the presence of executive dysfunction differentially impacted measures of verbal memory abilities. Older adults with impairment on selected measures of executive function (i.e., ED group) performed significantly worse than those without any demonstrated executive dysfunction (i.e., NoED) on the total learning and short delay recall components of the CVLT-II. However, on the other components of the CVLT-II and on the LM, there were no significant differences between those with and without executive

9 238 BRIAN L. BROOKS ET AL. dysfunction. Correlations performed between the executive function and verbal memory measures also further support these findings, with significant correlations found between executive measures and some of the CVLT-II indices, but not between the executive function measures and the LM measure. It has been suggested that differential performance in patients with executive dysfunction between the list-learning and story-recall measures arises from the differential requirements of the tasks for active learning and encoding (Tremont et al., 2000). The word list on the CVLT-II, comprised of four semantic categories with four words in each category, is presented to participants in a fashion that has been suggested to require active organization for optimal performance. In addition, previous research (cf. Vanderploeg et al., 1994) has suggested that attentional aspects of executive functions play a key role in optimal performance on list-learning measures. On the other hand, the LM test is presented in a structured story format, thus potentially reducing the need to organize the information for learning and recall by providing contextual cues. The prefrontal cortex influences the active organization of verbal information. The prefrontal cortex, one area of the brain important for successful performance on measures of executive function, plays a strategic and supervisory role in the encoding and retrieval of verbal episodic memories (Nyberg & Cabeza, 2000; Stuss & Levine, 2002). Neurological studies have supported the relationship between prefrontal functions (i.e., the cortical region implicated in executive functioning) and memory performance in neuropsychological assessments. Stuss and Levine (2002) review the neuroimaging literature and identify the dorsolateral prefrontal cortex (DLPFC) to be associated with several higher cognitive functions, including monitoring and manipulation of information in working memory, top-down guidance of attentional processes (e.g., switching attention, extra-dimensional set-shifting, selective attention, sustained attention), and control and direction of strategic processes of memory. The above neurophysiological relationship has been referred to as the frontal-striatal connection (Cummings, 1993), involving connections from the DLPFC, to the dorsolateral caudate nucleus, the lateral dorsomedial globus pallidus and the ventral anterior=medial dorsal thalamus, and then reciprocating back to the DLPFC. Aberrations in the frontal-striatal network are likely to appear as impairments on measures of executive functions (DeLong, 2000; Saper, Iverson, & Frackowiak, 2000), which in turn can negatively impact the encoding and retrieving of verbal information (Nyberg & Cabeza, 2000). Early in the progression of dementia and predictive of future decline from MCI to dementia, changes in the prefrontal lobes begin to manifest and show up as early decline in executive functions and coincide with impairments in verbal memory (Albert et al., 2001). It is plausible that these changes in the frontal-striatal network early in the progression of dementia negatively impact performance on the CVLT-II (Wheeler, Stuss, & Tulving, 1995), a task requiring active encoding and organization of material (Vanderploeg et al., 1994). However, changes in the frontal-striatal network may be less likely to impact performance on the LM due to the contextual cues provided by having the verbal information presented in story format. The present study lends support to Tremont et al. s (2000) findings that those persons with executive dysfunction perform significantly worse than those persons without executive dysfunction on indices of the CVLT, but not on the LM. In

10 EXECUTIVE DYSFUNCTION AND MEMORY 239 addition, the present study expands Tremont et al. s findings to a population of older adults with MCI and mild dementia. The present study found that those with significant executive dysfunction performed worse on the learning trials and short delay recall of the CVLT; however, there was not a significant difference on the long delay recall (although there was a trend towards a difference, p ¼.039) or recognition trials. One reason for this difference between the studies may be that among persons with MCI and mild dementia, memory impairment in terms of poor retrieval is a primary feature. The information may begin to degrade quickly after the short delay recall trial is completed; therefore, the 20-minute long delay may have a negative effect on the NoED group. As a result, even with better performance during the learning and short delay recall trials, the NoED group was eventually similar to the ED group, regardless of the quality of executive functioning. This possible explanation for the absence of a group difference after the long delay suggests that the prefrontal cortex provides a strategic, but not an absolute role in memory encoding and retrieval. Drawing comparisons between the present study and Tremont et al. (2000) should be done with some caution. For example, in addition to the Trails B, COWAT, and WCST-PR measures, Tremont et al. also included two measures to assess abstract verbal reasoning (WAIS-R Similarities subtest) and planning (the Boston Qualitative Scoring System for the Rey-Osterrieth Complex Figure). It is possible that those identified as having executive dysfunction in Tremont et al. are different from those in the present study. Also, an examination of the effect sizes for the mean scores shows that there is a similar effect size for CVLT total learning between the studies, but the effect sizes for CVLT LDFR, LMI, and LMII are substantially lower in the present study (.50,.40, and.07, respectively). Third, the present study utilized the more updated CVLT (2nd edition) and any differences in performance in relation to executive dysfunction between the 1st and 2nd editions of the CVLT are not known. One limitation of the Tremont et al. (2000) study, which the present study attempted to address, involved the method for establishing the groups. Tremont et al. grouped persons with zero or one impaired executive function measure (minimal ED) and grouped persons with two or three impaired executive function measures (significant ED), and then proceeded to analyze group differences across the measures. However, later analyses of the performance on the learning trials of the CVLT indicated that the sub-group without any impaired executive function measures differed significantly from the sub-group with one impaired executive function measure. The lack of homogeneity brings into question the validity of the analyses carried out comparing the minimal and significant ED groups. Unlike Tremont et al., the NoED group and the ED group in the present study were each homogeneous in terms of the independent variables. As well, the present study found that the presence of any executive dysfunction resulted in differential performance on the CVLT-II, but not the LM measure. A potential limitation of the present study is the possibility that presence of executive dysfunction is inextricably tied to degree of cognitive impairment. That the obtained group differences in learning and short delay recall are not just artifacts related to the degree of overall cognitive impairment, is suggested by the fact that the two executive function groups (i.e., NoED and ED) did not differ significantly in the

11 240 BRIAN L. BROOKS ET AL. number of persons who were patient controls, MCI or mild dementia; therefore, it is less likely that any difference between the groups on the CVLT-II scores is directly attributable to the presence of a dementing illness or overall cognitive dysfunction. As well, the groups did not differ on measures of cognitive functioning that are likely to decline with the progression of dementia (e.g., language, visual-spatial abilities), lending further support to the impact of executive dysfunction on the list-learning task. A further limitation of this study involves the measures that were selected. This study examined differential impairment using the Logical Memory subtest from the WMS-R, which is from 1987, and compared it to the CVLT-II, a measure that was updated in 2000 (Delis et al., 2000). However, it should be noted that the normative data for the LM came from the Mayo Older Americans Normative Study (MOANS; Ivnik et al., 1992), which provided more up-to-date norms than the WMS-R manual (Wechsler, 1987). Another limitation of the present study is related to the measures selected relates to the executive function measures. The present study selected common measures of executive function related to the neurobehavioural functions of the dorsolateral prefrontal cortex. The present study is unable to provide any information about the relation between verbal memory and impaired performance on measures that are related to compromised functioning in other regions of the prefrontal cortex. It is also possible that the selected executive function measures do not tap into the same processes that are involved with the CVLT-II Repetitions, Intrusions, and False Positives indices, and therefore would not be significantly different between the two groups. Further exploration of the relationship between these CVLT-II indices and executive functions in this population is needed. The present study examined whether the presence of executive dysfunction contributed to differential performance on measures of verbal memory. It was observed that those persons with impairment on measures of executive functions had lower levels of performance on the learning and short delay recall trials of the CVLT-II, a list learning measure, but did not have lower performance on the LM, a story recall measure, or other indices from the CVLT-II. This study is the first study known to the authors to examine executive dysfunction and differential verbal memory performance in a sample of older adults with suspected MCI or mild dementia. The results add to a recent literature in this area, contributing to our understanding of the component processes within memory testing, replicated now in several samples with diverse aetiologies. It appears that the presence of executive dysfunction may augment the impact of retrieval deficits in determining performance in learning and memory tests. The results suggest that the evaluation of memory performance in older adults with suspected dementia may benefit from a thorough evaluation of memory and executive functions using multiple measures. ACKNOWLEDGEMENTS This study was supported by a Mental Health Studentship from the Alberta Heritage Foundation for Medical Research to the first author. The authors thank Norma Cassini (former director) and Dr. Patricia Jean in seniors health at the Rockyview General Hospital in Calgary, Alberta.

12 EXECUTIVE DYSFUNCTION AND MEMORY 241 REFERENCES Albert, M. S., Moss, M. B., Tanzi, R., & Jones, K. (2001). Preclinical prediction of AD. Journal of the International Neuropsychology Society, 7, American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author. Benton, A. L. & Hamsher, K. (1976). Multilingual aphasia examination. Iowa City: University of Iowa Press. Cummings, J. L. (1993). Frontal-subcortical circuits and human behavior. Archives of Neurology, 50, Delis, D. C., Cullum, M., Butters, N., Cairns, P., & Prifitera, A. (1988). Wechsler Memory Scale Revised and California Verbal Learning Test: Convergence and divergence. The Clinical Neuropsychologist, 2, Delis, D. C., Kramer, J. H., Kaplan, E., & Ober, B. A. (1987). The California Verbal Learning Test manual. New York: The Psychological Corporation. Delis, D. C., Kramer, J. H., Kaplan, E., & Ober, B. A. (2000). The California Verbal Learning Test manual (2nd ed.). New York: The Psychological Corporation. DeLong, M. R. (2000). The basal ganglia. In E. R. Kandel, J. H. Schwartz, & T. M. Jessel (Eds.), Principles of neural science (4th ed., pp ). New York: McGraw-Hill. Fossati, P., Coyette, F., Ergis, A.-M., & Allilaire, J.-F. (2002). Influence of age and executive functioning on verbal memory of inpatients with depression. Journal of Affective Disorders, 68, Fox, L. S., Olin, J. T., Erblich, J., Ippen, C. G., & Scheider, L. S. (1998). Severity of cognitive impairment in Alzheimer s disease affects list learning using the California Verbal Learning Test (CVLT). International Journal of Geriatric Psychiatry, 13, Gladsjo, J. A., Schuman, C. C., Evans, J. D., Peavy, G. M., Miller, S. W., & Heaton, R. K. (1999). Norms for letter and category fluency: Demographic corrections for age, education, and ethnicity. Assessment, 6, Ivnik, R. J., Malec, J. F., Smith, G. E., Tangolos, E. G., & Petersen, R. C. (1996). Neuropsychological tests norms above age 55: COWAT, BNT, MAE Token, WRAT-R Reading, AMNART, STROOP, TMT, and JLO. The Clinical Neuropsychologist, 10, Ivnik, R. J., Malec, J. F., Smith, G. E., Tangalos, E. G., Petersen, R. C., Kokmen, E., et al. (1992). Mayo s older Americans normative studies: WAIS-R, WMS-R and AVLT norms for ages 56 through 97. The Clinical Neuropsychologist, 6, Kaplan, E. F., Goodglass, H., & Weintraub, S. (1978). The Boston Naming Test. Boston: E. Kaplan & H. Goodglass. Kongs, S. K., Thompson, L. L., Iverson, G. L., & Heaton, R. K. (2000). Wisconsin Card Sorting Test 64 card version, professional manual. Odessa, FL: Psychological Assessment Resources. Lezak, M. D. (1995). Neuropsychological assessment (3rd ed.). New York: Oxford University Press. Nyberg, L. & Cabeza, R. (2000). Brain imaging of memory. In E. Tulving & F. I. M. Craik (Eds.), Oxford handbook of memory (pp ). New York: Oxford University Press. Petersen, R. C., Doody, R., Kurz, A., Mohs, R. C., Morris, J. C., Rabins, P. V., et al. (2001). Current concepts in mild cognitive impairment. Archives of Neurology, 58, Petersen, R. C., Smith, G. E., Waring, S. C., Ivnik, R. J., Tangalos, E. G., & Kokmen, E. (1999). Mild cognitive impairment: Clinical characterization and outcome. Archives of Neurology, 56,

13 242 BRIAN L. BROOKS ET AL. Randolph, C., Gold, J. M., Kozora, E., Cullum, C. M., et al. (1994). Estimating memory function: Disparity of Wechsler Memory Scale Revised and California Verbal Learning Test indices in clinical and normal samples. The Clinical Neuropsychologist, 8, Reitan, R. M. & Wolfson, D. (1985). The Halstead-Reitan neuropsychological test battery. Tucson: Neuropsychology Press. Ringe, W. K., Frol, A. B., Saine, K. C., & Cullum, C. M. (1999). Organization and its relationship with other measures of executive functioning and memory. Archives of Clinical Neuropsychology, 14, 142. Saper, C. B., Iverson, S., & Frackowiak, R. (2000). Integration of sensory and motor function: The association areas of the cerebral cortex and the cognitive capabilities of the brain. In E. R. Kandel, J. H. Schwartz, & T. M. Jessel (Eds.), Principles of neural science (4th ed., pp ). New York: McGraw-Hill. Stuss, D. T. & Levine, B. (2002). Adult clinical neuropsychology: Lessons from studies of the frontal lobes. Annual Reviews in Psychology, 53, Tremont, G., Halpert, S., Javorsky, D. J., & Stern, R. A. (2000). Differential impact of executive dysfunction on verbal list learning and story recall. The Clinical Neuropsychologist, 14, Vanderploeg, R. D., Schinka, J. A., & Retzlaff, P. (1994). Relationships between measures of auditory verbal learning and executive functioning. Journal of Clinical and Experimental Neuropsychology, 16, Wechsler, D. (1987). Wechsler Memory Scale Revised. San Antonio, TX: The Psychological Corporation. Wechsler, D. (1997). WAIS-III and WMS-III technical manual. San Antonio, TX: The Psychological Corporation. Weinfurt, K. P. (1995). Multivariate analysis of variance. In L. G. Grimm & P. R. Yarnold (Eds.), Reading and understanding multivariate statistics (pp ). Washington, DC: American Psychological Association. Wheeler, M. A., Stuss, D. T., & Tulving, E. (1995). Frontal lobe damage produces episodic memory impairment. Journal of the International Neuropsychological Society, 1,