Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Jul 4.
Published in final edited form as: Clin Neuropsychol. 2025 Sep 13;40(7):1978–1995. doi: 10.1080/13854046.2025.2552279

Reliability and stability of cognitive intraindividual variability indices among cognitively unimpaired older adults

Sidonia E Compton a,b, Troy A Webber c,d,e, Steven Paul Woods d, Andrew M Kiselica f
PMCID: PMC13330730  NIHMSID: NIHMS2176073  PMID: 40944431

Abstract

Objective:

Cognitive intraindividual variability (IIV) dispersion measures within-person variability in performance across a neuropsychological test battery and is prognostic of structural and functional decline. Although evidence for the construct validity of cognitive dispersion has grown considerably, its utility in applied settings relies on its stability and reliability. Thus, we examined mean-level stability, test-retest reliability, and reliable change of several cognitive dispersion indices among cognitively unimpaired older adults.

Method:

Participants were 2224 robustly cognitively unimpaired older adults from the National Alzheimer’s Coordinating Center (NACC) Uniform Data Set 3.0. Intraindividual standard deviation (ISD) and coefficient of variation (CoV) dispersion indices were calculated for raw, scaled, and demographically-adjusted normed scores for the neuropsychological test battery at baseline and initial follow-up visit (Mdays = 424.99). Analyses included paired samples t-tests and generation of effect sizes for mean-level stability, correlations for test-retest reliability, and practice-adjusted reliable change indices (RCIs) for classifying individuals.

Results:

The mean-level change effect sizes for ISD and CoV scores were very small (ds < .06); only CoV showed significant improvements. Test-retest reliability was poor (Mr = .41) for cognitive dispersion indices compared to global composite scores (r = .82). RCIs suggested that normed score changes greater than 1.75 standard deviations were significant.

Conclusions:

Findings showed stable mean-level cognitive dispersion across 424.99 days; however, consistent with process-based neuropsychological test scores literature, test-retest reliability was poor, perhaps reflecting random measurement error among cognitively unimpaired individuals. Change in cognitive dispersion may inform clinical prognosis and serve as supplementary information within broader neuropsychological evaluations given reliability limitations.

Keywords: Intraindividual variability, cognitive dispersion, psychometrics, neuropsychological assessment, cognitive aging

Test-retest reliability of cognitive intraindividual variability indices in a sample of cognitively unimpaired older adults

Cognitive intraindividual variability (IIV) dispersion refers to within-person variability across a neuropsychological test battery (Costa et al., 2019). In brief, cognitive dispersion is most commonly calculated by generating the within-person standard deviation of a series of scores across a battery, which may be adjusted by mean-level performance. A degree of variability in cognitive performance is expected within a battery (Binder et al., 2009); however, increased cognitive dispersion is hypothesized to index executive control or the ability to marshal cognitive resources from various brain networks to consistently perform tasks over time (Kiselica, Karr, et al., 2024). Cognitive dispersion is strongly linked with executive functioning performance (Penheiro et al., 2024) and the integrity of the frontal lobe and connecting white matter structures (Karr et al., 2014; Kiselica, Karr, et al., 2024). Moreover, emerging evidence suggests that cognitive dispersion measures appear to have clinical utility. For example, these indices may be more sensitive to incipient cognitive decline than mean-level performance on neuropsychological test batteries (Bangen et al., 2019; Costa et al., 2019). Such findings have been reported in both cognitively unimpaired (Bangen et al., 2019; De Vito et al., 2024) and cognitively impaired older adult samples (Davis et al., 2023; Kälin et al., 2014; Webber et al., 2022). Additional findings underscore the role of cognitive dispersion in identifying incipient cognitive decline by linking cognitive dispersion to neuropathological correlates (e.g. amyloid-beta positivity and widespread network dysfunction across multiple brain networks; Holmqvist et al., 2023; Lin & McDonough, 2022). Cognitive dispersion has also been linked to declines in daily functioning, supporting the ecological validity of these measures (Scott et al., 2023; Webber, Lorkiewicz, Kiselica, et al., 2024).

While growing evidence highlights the potential utility of cognitive dispersion, there have been few investigations of the psychometrics of cognitive dispersion. This is an important area of work, as cognitive dispersion scores may also be influenced by factors such as effort, demographics, distraction, and characteristics of the underlying tests (Kiselica, Karr, et al., 2024; Webber, Lorkiewicz, Woods, et al., 2024). To address this gap, Kiselica, Kaser, et al. (2024) provided a regression-based approach for calculating normed scores for cognitive dispersion among older adults for application in clinical and research environments. These normed scores for cognitive dispersion were established using a normative sample and showed excellent validity for distinguishing cognitively unimpaired individuals from those with cognitive impairment due to probable Lewy body disease.

Thus, there is a need for more longitudinal research to understand temporal stability, test-retest reliability, and reliable change in dispersion measures. Such work is important for several reasons. First, analyzing mean level change would facilitate interpretation of sample level change over time in dispersion indices. Second, assessing test-retest reliability would increase our understanding of the degree to which dispersion scores are influenced by error, promoting reproducibility and improving interpretation of between group differences (Estabrook et al., 2012; Hopkins, 2000; Portney & Watkins, 2000). Finally, reliable change measures would enable evaluation of individual-level change in dispersion (Duff, 2012).

To our knowledge, there has only been one study on these topics among older adults. DesRuisseaux et al. (2024) reported on stability and change on one dispersion measure, the intraindividual standard deviation (ISD), in a sample of 238 patients with mild cognitive impairment assessed twice over an approximate 1.5-year interval. The test battery examined aspects of attention/processing speed, language, memory, visuospatial, and executive functioning abilities, and was used to generate standardized regression-based (SRB) z-scores for baseline and initial follow-up, accounting for across- and within-domain dispersion, across- and within-domain mean-level performance, and demographics (e.g. age and education). They reported 1) no significant mean level change in dispersion across time, 2) moderate test-retest reliability (r = .69), and 3) provided methods for calculating standardized regression-based change scores for the ISD.

This study marks a crucial first step toward understanding the temporal characteristics of one dispersion measure (the ISD). This work could be extended in several ways. First, there are other measures of dispersion, such as the coefficient of variation (CoV), that were not explored by DesRuisseaux et al. (2024). In the present study, we included both ISD and CoV metrics as CoV accounts for differences in variability that may be due to individual mean-level performance (Christensen et al., 2005). Second, these authors evaluated temporal characteristics of dispersion in a sample of patients with mild cognitive impairment (MCI), and findings need to be examined in a cognitively unimpaired population. Therefore, the present study included a sample of cognitively unimpaired older adults from the National Alzheimer’s Coordinating Center Uniform Data Set 3.0 who completed neuropsychological test batteries at two time points (baseline and first follow up visits). Third, DesRuisseaux et al. (2024) utilized raw ISD scores in their analyses. To examine temporal characteristics of normed dispersion scores, we calculated mean-level stability, test-retest reliability, and reliable change indices (RCIs) for normed scores for cognitive dispersion indices. Finally, as the prior study was based on a moderately sized clinical sample, the present study aimed to replicate their findings using a larger cohort derived from the NACC UDS 3.0 to improve the generalizability and reproducibility of findings. Additionally, we employed RCIs to determine whether changes in normed scores for cognitive dispersion over time are statistically significant beyond what could be attributed to measurement error. Based on prior research (DesRuisseaux et al., 2024), we predicted that there would be mean-level stability in dispersion and moderate test-retest reliability.

Methods

Procedures and participants

Longitudinal data from the Uniform Data Set (UDS) of the National Alzheimer’s Coordinating Center (NACC) were requested on August 29, 2024. The data file included UDS visits conducted between September 2005 and the data freeze in June 2024 from 39 Alzheimer’s Disease Research Centers (ADRCs). All participants provided written informed consent prior to enrollment and data collection. Detailed descriptions of participant recruitment and data collection procedures are reported elsewhere (Beekly et al., 2004; Besser et al., 2018; Morris et al., 2006; Weintraub et al., 2018).

Figure A1 outlines sample selection procedures in a Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) flow diagram (von Elm et al., 2007). Sample selection followed the procedures outlined by Kiselica, Kaser, et al. (2024) in developing normed dispersion scores for the UDS 3.0 neuropsychological battery. First, we included participants who completed the UDS 3.0 in-person at baseline and first follow up visits. Next, we restricted the sample to participants who were aged ≥ 50 and who primarily spoke English and completed tests administered in English. Subsequently, the sample was restricted to participants who were classified as cognitively unimpaired across all testing visits. In the current study, we defined cognitively unimpaired as obtaining a Clinical Dementia Rating (CDR®) Dementia Staging Instrument (Morris, 1993) global score of 0 and a NACC clinical diagnosis score of 1 (cognitively unimpaired). NACC clinical diagnosis classification at each visit was assessed by a clinician or consensus team using UDS form D1, available through the NACC website (https://naccdata.org/data-collection/forms-documentation/uds-3).

Participants with missing data on any cognitive measures or demographic variables of interest (age, sex, race/ethnicity, or education) were excluded. If participants were missing scores due to cognitive or behavior issues, they were included, and established winsorization methods were applied to their missing scores (Benge et al., 2022). Specifically, except for Trail Making Test A and B (TMT), a raw score of 0 was applied as this was the lowest possible score. For TMT-A and TMT-B, we assigned raw scores of 150 and 300, respectively, as these are interpretively the lowest possible scores for these measures. The final analytic sample comprised 2,224 older adults. Table A1 presents the sample’s demographic and clinical characteristics. Data on mood symptoms (Geriatric Depression Scale [GDS]; Yesavage, 1988) and the presence of a history of medical conditions, determined by the clinician’s best judgment following the medical interview at baseline using UDS form A5 available through the NACC website, were included to further characterize the sample. Table S1 summarizes the distribution of race and ethnic groups in the sample.

Cognitive measures

A detailed description of the Uniform Data Set 3.0 Neuropsychological Battery (UDS3NB) can be found elsewhere (Besser et al., 2018; Weintraub et al., 2018). For the purposes of the current study, we selected six tests, comprising 12 core scores, from the UDS3NB based on prior validation research (Webber et al., 2022; Webber, Lorkiewicz, Kiselica, et al., 2024) and the availability of demographically-adjusted dispersion scores for this specific battery (Kiselica, Kaser, et al., 2024). The 12 core scores were as follows: (1) the Craft Story verbatim recall scores, which measure immediate and delayed recall for verbal, contextual information (Craft et al., 1996); (2) the Benson Figure, which assesses visuoconstruction abilities and delayed visual recall (Possin et al., 2021); (3) Number Span Forward and Backward Test, which evaluates attention and working memory with digit repetition tasks; (4) the Multilingual Naming Test (MINT), which assesses confrontation naming (Gollan et al., 2012; Ivanova et al., 2013); (5) Letter (F- and L-) and semantic (animals and vegetables) fluency tasks, which measure verbal fluency; and (6) Trail Making Test A and B, which evaluates simple number sequencing and alternating number-letter sequencing, respectively (Partington & Leiter, 1949).

Cognitive dispersion indices

We calculated two cognitive dispersion indices: intraindividual standard deviation (ISD; Holtzer et al., 2008; Morgan et al., 2012; Webber et al., 2022; Webber, Lorkiewicz, Kiselica, et al., 2024) and coefficient of variation (CoV) for raw, scaled, and demographically-adjusted normed scores at the baseline and first follow up visits. To calculate ISD, we first computed demographically-adjusted (age, sex, education) z-scores following procedures established by Weintraub et al. (2018). Next, z-scores were converted into T-scores with a mean of 50 and SD of 10 using the following equation: (10*z-score) + 50. We computed ISD by taking the square root of the aver-age squared deviation of each test’s T-score from each participant’s UDS3NB normative mean score (i.e. the average of the 12 demographically-adjusted T-scores). To calculate CoV, we divided ISD by the participant’s UDS3NB normative mean score to control for between-subjects confounding effects of global cognitive abilities (Stawski et al., 2019). In addition, raw unadjusted z-scores were converted to scaled scores with a mean of 10 and SD of 3 using normative tables provided by Kiselica, Kaser, et al. (2024). For scaled scores, lower values represented higher cognitive dispersion (i.e. greater cognitive impairment; Hilborn et al., 2009) since lower scaled scores typically imply more cognitive impairment in the context of neuropsychological assessments.

Cognitive dispersion indices were further transformed into demographically-adjusted z-scores by utilizing the regression-based formula derived by Kiselica, Kaser, et al. (2024). Accounting for demographic influences, the regression equation includes age (years), race/ethnicity (dichotomized as White/non-Hispanic and non-White/Hispanic), and education (years, with an upper limit of 20 years), but not sex. Mathematical details for the computation of the standardized regression-based formula are described in Kiselica, Kaser, et al. (2024). We applied the following formulas to the obtained scaled scores to calculate the demographically-adjusted z-scores.

For demographically-adjusted ISD:

AdjustedISDz-score=(ISDscaledscore-(13.98+−0.82*(race/ethnicity+1)+−2.67*(education/10)−0.5+−2.41*(age/100)3))/2.96

For demographically-adjusted CoV:

AdjustedCoVz-score=(CoVscaledscore-(10.99+−1.55*(race/ethnicity+1)+−2.22*(age/100)3+−15.14*((age/100)3*log(age/100)))/2.91.

Data analyses

Mean-level stability and test-retest reliability

Skewness values (ranging from −0.032 to .094) and kurtosis values (ranging from −0.146 to −0.009) suggested that all scaled score and demographically-adjusted cognitive dispersion indices were normally distributed (Hair et al., 2010). Visual inspection of histograms and box plots suggested mild deviations from normality, which is accounted for by the large sample size.

We performed paired samples t-tests to examine mean differences and paired samples correlations for test-retest reliability. For comparison purposes, we conducted analyses for mean T-scores (global composite scores), in addition to raw scores, scaled scores, and demographically-adjusted z-scores for dispersion. As a supplemental analysis, we investigated the differences in mean stability and test-retest correlation between age groups via paired samples t-tests and correlations with four baseline age-stratified groups: (1) 50–59 years, (2) 60–69 years, (3) 70–79 years, and (4) ≥ 80 years. Cohen’s d was utilized to measure effect sizes. Following conventional definitions, Cohen’s d was defined as very small (≤ 0.19), small (0.20–0.49), medium (0.59–0.79), large (0.80–1.19), and very large (≥ 1.20) (Sawilowsky, 2009). Test-retest correlation values were interpreted as poor (r < 0.50), moderate (r = 0.50 to 0.75), good (r = 0.75 to 0.90), and excellent (r ≥ 0.90) (Portney & Watkins, 2000). To account for potential temporal effects, we also report test-retest estimates stratified by the following test-retest time intervals: (1) 6 to 12 months, (2) 12 to 18 months, and (3) 18 to 24 months. Of note, the mean number of days between visits was 424.99 (SD = 86.81, range: 185 to 730).

Reliable change indices (RCI) analysis

To examine associations between difference scores and demographic variables, we calculated Pearson correlations for continuous demographic variables and bi-point serial correlations for categorical demographic variables (see Table S3). No demographic variables were significantly associated with change in dispersion scores. Therefore, we opted not to use a standardized regression-based change approach for reliable change. Rather, we employed the easier-to-understand and less computationally intensive method of practice-adjusted reliable change indices with 90% confidence intervals (Chelune et al., 1993).

Results

Table A2 summarizes findings from paired samples t-tests and paired samples correlations. For raw indices, ISD scores were not significantly different, t(2223) = 0.54, p = .587, d = 0.01, while a significant difference was observed showing a very small improvement in raw CoV (t(2223) = 2.06, p = .039, d = 0.04). Similarly, for the scaled scores, there was no significant difference for ISD, t(2223) = −0.60, p = .548, d = −0.01, but a significant difference was observed showing a very small improvement in scaled CoV, t(2223) = −2.52, p = .012, d = −0.05. Further, there were no significant differences in demographically-adjusted ISD, t(2223) = −1.17, p = .244, d = −0.03; however, we observed a significant difference showing a very small improvement in demographically-adjusted CoV, t(2223) = −3.03, p = .002, d = −0.06.

Regarding test-retest reliability, a significant paired samples correlation was observed for raw ISD (r = .41, p < .001) and CoV (r = .51, p < .001). Results showed significant correlations for scaled ISD (r = .37, p < .001) and CoV (r = .43, p < .001), as well as demographically-adjusted ISD (r = .35, p < .001) and CoV (r = .39, p < .001). Similar findings emerged for stratified analyses for test-retest intervals (see Table A2). Test-retest correlations remained in the poor to moderate range when stratified by age bands (see Table S2).

Reliable change indices (RCI)

Table A3 provides the means and standard deviations of the simple difference scores for the various IIV metrics between baseline and the first follow-up visit, along with 90% confidence intervals for determining individual-level change. For example, an individual would have to demonstrate a CoV scaled score decline of > −5.12 or an increase in CoV scaled score of > 5.48 to be considered as having statistically significant change based on these data. Similarly, an individual would need to demonstrate a demographically adjusted ISD z-score decline of > −1.88 or an increase in a demographically adjusted ISD z-score of > 1.94 to be considered as having statistically significant change according to these findings. Based on this, 117 (5.3%) participants demonstrated significant decline on the global composite score from baseline to first follow up. Among stratified test-retest intervals, 29 (5.7%), 77 (5.0%), and 11 (6.1%) participants demonstrated significant decline on the global composite score for the (1) 6 to 12 months, (2) 12 to 18 months, and (3) 18 to 24 months intervals, respectively. Note that the SD values can be used in conjunction with the M difference scores to generate more liberal 68% confidence intervals. Results were also generated with 68% confidence intervals as a supplemental analysis (see Table S4).

Discussion

Mounting evidence suggests cognitive dispersion holds promise as a sensitive indicator of neuropsychological impairment. Although numerous validation studies lend support to the construct validity of cognitive dispersion, and recent efforts have developed methods for considering cognitive dispersion in applied settings (e.g. Kiselica, Kaser, et al., 2024), few studies have published data on the corresponding psychometrics of these indicators, particularly test-retest reliability and stability. To this end, the current study utilized a large sample of robustly cognitively unimpaired older adults to examine the temporal stability, test-retest reliability, and reliable change of six cognitive dispersion measures across 6–24 month test-retest intervals. Consistent with hypotheses and previous research (DesRuisseaux et al., 2024), sample-level results indicate that cognitive dispersion is largely stable across a 6–24 month interval for robustly cognitively unimpaired older adults. At the individual change level, cognitive dispersion had poor test-retest reliability and correspondingly wide RCI confidence intervals in this sample, which is inconsistent with study hypotheses and previous research (DesRuisseaux et al., 2024).

Consistent with what was reported by DesRuisseaux et al. (2024) for raw ISD in a sample of older adults with mild cognitive impairment tested twice over a 1.5-year interval, the current study shows that all six cognitive IIV dispersion metrics (i.e. raw ISD, raw CoV, scaled ISD, scaled CoV, demographically adjusted z-score ISD, demographically adjusted z-score CoV) exhibited gross temporal stability across a 6–24 month interval. Although some analyses showed significant change in the cognitive dispersion metrics over time, the effect sizes were negligible to very small (absolute values of d = .01 to .06), suggesting that the statistically significant differences may have been primarily driven by the very large sample size. As such, our results align with was reported by DesRuisseaux et al. (2024), and facilitate understanding of sample level change in cognitive dispersion, or lack thereof, among robustly cognitively healthy older adults. The absence of notable group-level change in cognitive dispersion in a sample of stable, healthy older adults supports the psychometrics of these indicators.

However, the analyses of individual-level changes in cognitive dispersion raised some concerns. In contrast to what was hypothesized and reported by DesRuisseaux et al. (2024; r = .69) in people with MCI, the test-retest reliability of cognitive dispersion metrics was poor (Mr = .41) in our sample of robustly cognitively unimpaired older adults. The substantially lower test-retest reliability estimate of cognitive dispersion metrics compared to the findings of DesRuisseaux et al. (2024) may reflect an interaction of psychometric and sample-specific considerations. Specifically, what is being measured in people with MCI may be quite different than what is being measured by in robustly cognitively unimpaired samples. To elaborate, variability within a neuropsychological test battery is common, with multiple normatively low scores across a battery frequently observed among cognitively unimpaired examinees (Binder et al., 2009; Donnell et al., 2011; Kiselica et al., 2020). At lower levels, this variability is partially (and in some cases entirely) explained by measurement error and considered “normal psychometric variability,” (Bangen et al., 2019; Kiselica et al., 2020). Normal psychometric variability is more likely to be observed among individuals with above-average general cognitive abilities, who are less prone to floor effects (i.e. below average performance among individuals with lower general cognitive abilities), with fluctuations in performance typically manifesting as lower scores (Hill et al., 2013). Variability within a neuropsychology test battery could also arise as a consequence of non-pathological age-related changes, such as reduced neural integrity (Lee & Kim, 2022), as some studies demonstrate decline in serial assessments among cognitively unimpaired older adults (Malek-Ahmadi et al., 2018). Alternatively, at extreme levels, it can be pathological and predictive of incipient cognitive decline (Bangen et al., 2019; Kiselica et al., 2020). Although not clearly evident in our study, it is possible that some cognitively unimpaired participants are demonstrating pre-clinical decline given that intraindividual variability can predict progression to mild cognitive impairment or dementia (Anderson et al., 2016; Gleason et al., 2018).

Notably, when using robustly cognitively unimpaired samples (i.e. cognitively unimpaired at baseline and first follow up), the sample selection process virtually eliminates the potential that variability at baseline is pathological. In other words, variability at baseline in our sample was random measurement error, not caused by a systematic underlying deficit in the ability to marshal cognitive resources to consistently perform tasks over time. In turn, extreme levels of variability at baseline would exhibit regression to the mean (Donnell et al., 2011) and show weak correlations with variability at first follow up. In contrast, the variability observed by DesRuisseaux et al. (2024) in MCI was not random, but more likely due to an underlying brain impairment. As such, cognitive dispersion in robustly cognitively healthy individuals may represent no more than measurement error, which by definition is less likely to track reliably over time in individual cases.

Relatedly, the substantially lower test-retest reliability for the cognitive dispersion metrics (Mr = .41) compared to the overall test battery mean (i.e. global composite score; r = .82) may reflect 1) how error is capitalized in the formulae and 2) regression to the mean. The formulae for cognitive dispersion exponentially amplifies measurement error/variability because the individual test score deviations from the overall test battery mean are squared. Considering that variability in neuropsychological test scores for robustly cognitively unimpaired individuals is more so due to random measurement error, which is subject to regression to the mean on the second measurement occasion, the exponentially amplified measurement error is unlikely to exhibit high test-retest reliability in this sample. With respect to the overall test battery mean, measurement error/variability is attenuated when the mean test battery score is calculated because regression to the mean causes centralization of error around the individual’s global ability level. Thus, the contribution of error from one or two unusually high or low scores is quite limited, which is reflected in Table A2 by the relatively small standard deviations (SDs) around the overall test battery mean (SDs are ~9% of the mean) but large SDs around the cognitive dispersion metrics (SDs are ~30% of the mean). As such, the attenuated measurement error for the overall test battery mean shows a higher degree of test-retest reliability because the contribution of error from any individual test score is attenuated, not exponentiated.

In a similar fashion, the discrepant findings of strong temporal stability at the sample level (as evidenced by negligible to very small Cohen’s d effect sizes) but poor test-retest reliability at the individual level (as evidenced by weak r effect sizes) may be explained by the fact that variability in the cognitive dispersion metrics represent measurement error in this robustly cognitively unimpaired sample. At the group level (i.e. estimates of temporal stability), variability/measurement error is attenuated because the error is subject to regression to the mean. Given that the variability/measurement error is randomly distributed across individuals in the sample, the group level change in error over a 6–24 month period is limited. In contrast, the test-retest reliability coefficients estimate stability at the individual level. Assuming that cognitive dispersion is disproportionately measurement error in robustly cognitively unimpaired individuals, measurement error at one occasion would be subject to regression to the mean at the other occasion, resulting in low stability at the individual level (Donnell et al., 2011).

The mixed stability and reliability findings for cognitive dispersion from this study warrant contextualization with the broader literature on the validity of cognitive dispersion. Specifically, both the current study and work by DesRuisseaux et al. (2024) on the psychometrics of cognitive dispersion suggest that it has good stability, but findings are mixed with regard to its test-retest reliability. There are, of course, psychometric precedents in the literature on higher-order cognition for which we observe poor reliability vis-a-vis adequate-to-strong evidence of validity. For example, some aspects of executive functions (e.g. novel problem solving, errors) and episodic memory (e.g. intrusions and perseverations) demonstrate poorer stability and/or reliability (e.g. Woods et al., 2005), but nevertheless show evidence of validity and potential clinical utility. We see a similar picture emerging for cognitive dispersion, which has strong evidence for its construct validity and potential clinical value, with convergent studies supporting its potential prognostic value in signaling incipient neurocognitive decline (Bangen et al., 2019; Costa et al., 2019), ability to distinguish between healthy samples and persons with central nervous system (CNS) compromise (Grewal et al., 2023), associations with biomarkers (e.g. neuroimaging; Karr et al., 2014; Kiselica et al., 2024) and executive functions (Penheiro et al., 2024), and sensitivity to everyday functioning problems (Kiselica et al., 2024). Moreover, it is important to recognize the lack of consensus regarding the most robust practices and metrics to use when evaluating intraindividual variability. For instance, ISD, though computationally simple and interpretable, is limited by its correlation with mean level performance (Estabrook et al., 2012). By contrast, although CoV accounts for mean level performance, future research is needed to explore how inflation of variability for below average individuals and the high degree of covariation shared between ISD and overall mean performance impacts the interpretation of cognitive dispersion findings (Del Bene et al., n.d.), particularly their reliability and stability. Indeed, we are still at the earliest stages of understanding the psychometrics of cognitive dispersion.

There were several limitations that should be considered when interpreting our findings, which primarily includes constraints on generality. Perhaps chief among these limitations is that these findings were observed in a relatively racially homogenous, older adult, and highly educated sample. Although the demographically adjusted normative data accounts for the some of the impact of these characteristics on the psychometrics of the cognitive dispersion metrics, replication in younger adult (< 50), less educated, and racially heterogenous samples would increase the external validity of these findings and promote use of cognitive dispersion metrics in more diverse applied settings. This suggestion is further supported by the test-retest data presented in Table S2, which shows that the reliability of cognitive dispersion metrics may not be entirely immune from the effects of age, even when using demographically adjusted normative scores. Similarly, these findings should be interpreted in the context of the 6–24 month test-retest interval. As shown in Table A3, the test-retest reliability of the cognitive dispersion metrics trended better for those re-tested in the 6–12 month band and 13–18 month band than those in the 19–24 month band. Future research would benefit from examining the psychometric properties of cognitive dispersion metrics for other test-retest intervals (e.g. 1–30 days, 1–5 months, 2–3 years) to more fully understand their measurement properties and facilitate use in applied settings. Further, as the current study followed the computational procedures for calculating normed dispersion scores for the Uniform Data Set 3.0 neuropsychological battery derived by Kiselica, Kaser, et al., 2024, findings may not generalize to test batteries and may differ based on the number of measures used, scores derived from each measure, and cognitive domains assessed, although no consensus yet exists in the literature on the optimal parameters needed to accurately cognitive dispersion (Del Bene et al., n.d.).

Finally, it is critical that these data are interpreted within the context of the sample’s cognitive impairment status, particularly considering that we utilized robustly cognitively unimpaired individuals. To further test the hypothesis that cognitive dispersion represents measurement error in robustly cognitively healthy individuals but represents a signal of pathology in individuals that may not be robustly cognitively healthy (e.g. individuals with subjective cognitive complaints only, MCI, or dementia), future research would benefit from examining the test-retest reliability of cognitive dispersion metrics in samples of individuals with extremely subtle and/or extremely severe cognitive decline. Relatedly, although our methods foster confidence in the cognitive impairment status of participants at baseline, it remains possible that participants developed cognitive impairment in the interval but were nevertheless diagnosed as cognitively unimpaired at the follow-up. Indeed, a small proportion of our sample (5.3%) demonstrated significant decline on the global composite score between baseline and first follow up; however, results may not be indicative of neuropathology or a neurocognitive disorder. Moreover, without comparing cognitive abilities to gold standard biomarkers, we cannot definitively rule out the presence of brain pathology. In other words, our procedures do not eliminate the possibility that participants were in fact cognitively impaired at follow-up, which may partially explain the poor test-retest reliability of our cognitive dispersion metrics. In contrast, the findings reported by DesRuisseaux et al. (2024) were observed in individuals with MCI, which presumably reflected a stable or progressive underlying brain condition and increases confidence that the cognitive dispersion observed within their sample was pathological, rather than measurement error. Perhaps, clinical samples demonstrate more reliable and stable intraindividual variability due to the underlying pathology; however, future work is needed in this area. Limitations notwithstanding, these data represent a critical step towards understanding the construct validity of cognitive dispersion metrics and validating cognitive dispersion metrics for future use in applied settings.

Supplementary Material

Supplemental appendix

Supplemental data for this article can be accessed online at https://doi.org/10.1080/13854046.2025.2552279.

Acknowledgements

NACC data are contributed by the NIA-funded ADRCs: P30 AG062429 (PI James Brewer, MD, PhD), P30 AG066468 (PI Oscar Lopez, MD), P30 AG062421 (PI Bradley Hyman, MD, PhD), P30 AG066509 (PI Thomas Grabowski, MD), P30 AG066514 (PI Mary Sano, PhD), P30 AG066530 (PI Helena Chui, MD), P30 AG066507 (PI Marilyn Albert, PhD), P30 AG066444 (PI David Holtzman, MD), P30 AG066518 (PI Lisa Silbert, MD, MCR), P30 AG066512 (PI Thomas Wisniewski, MD), P30 AG066462 (PI Scott Small, MD), P30 AG072979 (PI David Wolk, MD), P30 AG072972 (PI Charles DeCarli, MD), P30 AG072976 (PI Andrew Saykin, PsyD), P30 AG072975 (PI Julie A. Schneider, MD, MS), P30 AG072978 (PI Ann McKee, MD), P30 AG072977 (PI Robert Vassar, PhD), P30 AG066519 (PI Frank LaFerla, PhD), P30 AG062677 (PI Ronald Petersen, MD, PhD), P30 AG079280 (PI Jessica Langbaum, PhD), P30 AG062422 (PI Gil Rabinovici, MD), P30 AG066511 (PI Allan Levey, MD, PhD), P30 AG072946 (PI Linda Van Eldik, PhD), P30 AG062715 (PI Sanjay Asthana, MD, FRCP), P30 AG072973 (PI Russell Swerdlow, MD), P30 AG066506 (PI Glenn Smith, PhD, ABPP), P30 AG066508 (PI Stephen Strittmatter, MD, PhD), P30 AG066515 (PI Victor Henderson, MD, MS), P30 AG072947 (PI Suzanne Craft, PhD), P30 AG072931 (PI Henry Paulson, MD, PhD), P30 AG066546 (PI Sudha Seshadri, MD), P30 AG086401 (PI Erik Roberson, MD, PhD), P30 AG086404 (PI Gary Rosenberg, MD), P20 AG068082 (PI Angela Jefferson, PhD), P30 AG072958 (PI Heather Whitson, MD), P30 AG072959 (PI James Leverenz, MD).

Funding

Dr. Kiselica is supported by a career development award from the National Institute on Aging (NIA) of the National Institutes of Health under Award Number U54AG063546, which funds NIA Imbedded Pragmatic Alzheimer’s Disease and AD-Related Dementias Clinical Trials Collaboratory (NIA IMPACT Collaboratory). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The NACC database is funded by NIA/NIH Grant U24 AG072122.

Appendix

Figure A1.

Figure A1.

Open in a new tab

STROBE diagram for sample selection.

Note. STROBE = Strengthening the Reporting of Observational Studies in Epidemiology. UDS3NB = Uniform Data Set 3.0 Neuropsychological Battery. CDR® = Clinician Dementia Rating

Table A1.

Demographic and clinical characteristics descriptives.

Variable M (SD)
Baseline age (years) 69.25 (7.34)
Education (years) 16.63 (2.28)
Baseline GDS 1.23 (4.79)
First Follow-up GDS 1.31 (5.48)
Follow-up interval (Days) 424.99 (86.81)
n (%)
Sex
 Male 784 (35.3)
 Female 1440 (64.7)
Race/ethnicity
 White/non-Hispanic 1669 (75.0)
 non-White/Hispanic 555 (25.0)
Baseline Medical Conditions
 Diabetes (Recent/active or Remote/inactive) 255 (11.4)
 Hypertension (Recent/active or Remote/inactive) 934 (42.0)
 Hypercholesterolemia (Recent/active or Remote/inactive) 1090 (49.0)
 Vitamin B12 deficiency (Recent/active or Remote/inactive) 128 (5.8)
 Thyroid disease (Recent/active or Remote/inactive) 432 (19.4)
 Arthritis (Recent/active or Remote/inactive) 1217 (54.7)
 Incontinence – urinary (Recent/active or Remote/inactive) 249 (11.2)
 Incontinence – bowel (Recent/active or Remote/inactive) 56 (2.5)
 Sleep Apnea (Recent/active or Remote/inactive) 382 (17.2)
 RBD (Recent/active or Remote/inactive) 35 (1.5)
 Hyposomnia/insomnia (Recent/active or Remote/inactive) 318 (14.3)
Open in a new tab

Note. Education was recoded so that 20 years was the highest level of education attained. GDS = Geriatric Depression Scale (Yesavage, 1988), BMI = Body mass index, RBD = REM sleep behavior disorder. Recent/active refers to if the medical condition happened within the last year or still requires active management and is consistent with information obtained from the subject and co-participant interview. Remote/inactive refers to if the medical condition existed or occurred in the past (more than one year ago) but was resolved or there is no treatment currently under way.

Table A2.

Mean change and test-retest correlations for IIV variables across the baseline and first follow up evaluations.

IIV Variable Time 1 M (SD) Time 2 M (SD) t, p d rm
Global Composite 50.09 (4.70) 50.70 (4.75) −9.98, <.001 −0.21
Raw ISD 7.82 (2.44) 7.79 (2.39) 0.54, .587 0.01
Raw CoV 0.16 (0.06) 0.16 (0.06) 2.06, .039 0.04
Scaled ISD 10.26 (3.04) 10.30 (3.04) −0.60, .548 −0.01
Scaled CoV 10.39 (3.00) 10.56 (3.01) −2.52, .012 −0.05
Dem. Adj. ISD z-score 0.07 (1.02) 0.10 (1.02) −1.17, .244 −0.03
Dem. Adj. CoV z-score 0.12 (1.00) 0.19 (1.01) −3.03, .002 −0.06
IIV variable Overall r, p (n = 2224) 6–12 months r, p (n = 512) 13–18 months r, p (n = 1531) 19–24 months r, p (n = 181)
Global Composite .82, <.001 .85, <.001 .81, <.001 .81, <.001
Raw ISD .41, <.001 .55, <.001 .39, <.001 .20, .008
Raw CoV .51, <.001 .64, <.001 .48, <.001 .30, <.001
Scaled ISD .37, <.001 .43, <.001 .35, <.001 .32, <.001
Scaled CoV .43, <.001 .51, <.001 .40, <.001 .38, <.001
Dem. Adj. ISD z-score .35, <.001 .42, <.001 .33, <.001 .30, <.001
Dem. Adj. CoV z-score .40, <.001 .49, <.001 .37, <.001 .32, <.001
Open in a new tab

Note. drm = Cohen’s d for repeated measures; IIV = intraindividual variability. ISD = Intraindividual standard deviation; CoV = coefficient of variation; Dem. = demographically; Adj. = adjusted.

Table A3.

Descriptive statistics for difference scores between baseline and first follow up at a 90% confidence interval.

Variable M SD a 90% CI
Global Composite difference 0.61 2.86 −4.10, 5.32
Raw ISD difference −0.03 2.61 −4.32, 4.26
Raw CoV difference 0.00 0.06 −0.10, 0.10
Scaled ISD difference 0.04 3.42 −5.59, 5.67
Scaled CoV difference 0.18 3.22 −5.12, 5.48
Dem. adj. ISD z-score difference 0.03 1.16 −1.88, 1.94
Dem. adj. CoV z-score difference 0.07 1.11 −1.76, 1.90
Open in a new tab

Note. M is Mean; SD is Standard Deviation; ISD = intraindividual standard deviation; CoV = coefficient of variation; Dem. = demographically; Adj. = adjusted.

a

Represents the 68% confidence interval (+/− practice effect).

Footnotes

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available by request through the National Alzheimer’s Coordinating Center (NACC) at https://naccdata.org/requesting-data/data-request-process.

References

  1. Anderson ED, Wahoske M, Huber M, Norton D, Li Z, Koscik RL, Umucu E, Johnson SC, Jones J, Asthana S, & Gleason CE, Alzheimer’s Disease Neuroimaging Initiative. (2016). Cognitive variability-A marker for incident MCI and AD: An analysis for the Alzheimer’s Disease Neuroimaging Initiative. Alzheimer’s & Dementia: Diagnosis, Assessment & Disease Monitoring, 4(1), 47–55. 10.1016/j.dadm.2016.05.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bangen KJ, Weigand AJ, Thomas KR, Delano-Wood L, Clark LR, Eppig J, Werhane ML, Edmonds EC, & Bondi MW (2019). Cognitive dispersion is a sensitive marker for early neurodegenerative changes and functional decline in nondemented older adults. Neuropsychology, 33(5), 599–608. 10.1037/neu0000532 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Beekly DL, Ramos EM, van Belle G, Deitrich W, Clark AD, Jacka ME, & Kukull WA, NIA-Alzheimer’s Disease Centers. (2004). The National Alzheimer’s Coordinating Center (NACC) Database: An Alzheimer disease database. Alzheimer Disease and Associated Disorders, 18(4), 270–277. [PubMed] [Google Scholar]
  4. Benge JF, Artz JD, & Kiselica AM (2022). The ecological validity of the Uniform Data Set 3.0 neuropsychological battery in individuals with mild cognitive impairment and dementia. The Clinical Neuropsychologist, 36(6), 1453–1470. 10.1080/13854046.2020.1837246 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Besser L, Kukull W, Knopman DS, Chui H, Galasko D, Weintraub S, Jicha G, Carlsson C, Burns J, Quinn J, Sweet RA, Rascovsky K, Teylan M, Beekly D, Thomas G, Bollenbeck M, Monsell S, Mock C, Zhou XH, … Morris JC, Neuropsychology Work Group, Directors, and Clinical Core leaders of the National Institute on Aging-funded US Alzheimer’s Disease Centers. (2018). Version 3 of the National Alzheimer’s Coordinating Center’s Uniform Data Set. Alzheimer Disease and Associated Disorders, 32(4), 351–358. 10.1097/WAD.0000000000000279 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Binder LM, Iverson GL, & Brooks BL. (2009). To err is human: “abnormal” neuropsychological scores and variability are common in healthy adults. Archives of Clinical Neuropsychology: The Official Journal of the National Academy of Neuropsychologists, 24(1), 31–46. 10.1093/arclin/acn001 [DOI] [PubMed] [Google Scholar]
  7. Chelune GJ, Naugle RI, Lüders H, Sedlak J, & Awad IA. (1993). Individual change after epilepsy surgery: Practice effects and base-rate information. Neuropsychology, 7(1), 41–52. 10.1037/0894-4105.7.1.41 [DOI] [Google Scholar]
  8. Christensen H, Dear KB, Anstey KJ, Parslow RA, Sachdev P, & Jorm AF (2005). Within-occasion intraindividual variability and preclinical diagnostic status: Is intraindividual variability an indicator of mild cognitive impairment? Neuropsychology, 19(3), 309–317. 10.1037/0894-4105.19.3.309 [DOI] [PubMed] [Google Scholar]
  9. Costa AS, Dogan I, Schulz JB, & Reetz K. (2019). Going beyond the mean: Intraindividual variability of cognitive performance in prodromal and early neurodegenerative disorders. The Clinical Neuropsychologist, 33(2), 369–389. 10.1080/13854046.2018.1533587 [DOI] [PubMed] [Google Scholar]
  10. Craft S, Newcomer J, Kanne S, Dagogo-Jack S, Cryer P, Sheline Y, Luby J, Dagogo-Jack A, & Alderson A. (1996). Memory improvement following induced hyperinsulinemia in Alzheimer’s disease. Neurobiology of Aging, 17(1), 123–130. 10.1016/0197-4580(95)02002-0 [DOI] [PubMed] [Google Scholar]
  11. Davis JJ, Sivaramakrishnan A, Rolin S, & Subramanian S. (2023). Intra-individual variability in cognitive performance predicts functional decline in Parkinson’s disease (pp. 1–8). Adult. 10.1080/23279095.2022.2157276 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. De Vito AN, Ju CH, Lee SY, Cohen AK, Trofimova AD, Liu Y, Eichten A, & Hughes A. (2024). Cognitive dispersion is related to subtle objective daily functioning changes in older adults with and without cognitive impairment. Alzheimer’s & dementia (Amsterdam, Netherlands), 16(1), e12539. 10.1002/dad2.12539 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Del Bene VA, Aita S, Fonseca LM, Borgogna N, Buchholz AS, Woods SP, Schretlen DJ, Kiselica AM, Webber T, Schmitter-Edgecombe M, DesRuisseaux LA, Merritt VC, Thaler NS, Bangen KJ, Vance DE, Deboeck PR, & Renteria MA Hill BD (n.d.). Cognitive intra-individual variability dispersion as an emerging measure of neuropsychological inference: A narrative review of its history, methodology. empirical support, future directions, and recommendations for best practices. [Manuscript submitted for publication]. Department of Neurology, University of Alabama at Birmingham. [Google Scholar]
  14. DesRuisseaux LA, Guevara JE, & Duff K. (2024). Examining the stability and predictive utility of across- and within-domain intra-individual variability in mild cognitive impairment. Archives of clinical neuropsychology: The official journal of the National Academy of Neuropsychologists, acae054. Advance online publication. 10.1093/arclin/acae054 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Donnell AJ, Belanger HG, & Vanderploeg RD (2011). Implications of psychometric measurement for neuropsychological interpretation. The Clinical Neuropsychologist, 25(7), 1097–1118. 10.1080/13854046.2011.599819 [DOI] [PubMed] [Google Scholar]
  16. Duff K. (2012). Evidence-based indicators of neuropsychological change in the individual patient: Relevant concepts and methods. Archives of Clinical Neuropsychology: The Official Journal of the National Academy of Neuropsychologists, 27(3), 248–261. 10.1093/arclin/acr120 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Estabrook R, Grimm KJ, & Bowles RP (2012). A Monte Carlo simulation study of the reliability of intraindividual variability. Psychology and Aging, 27(3), 560–576. 10.1037/a0026669 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Gleason CE, Norton D, Anderson ED, Wahoske M, Washington DT, Umucu E, Koscik RL, Dowling NM, Johnson SC, Carlsson CM, & Asthana S, Alzheimer’s Disease Neuroimaging Initiative. (2018). Cognitive variability predicts incident Alzheimer’s disease and mild cognitive impairment comparable to a cerebrospinal fluid biomarker. Journal of Alzheimer’s Disease: JAD, 61(1), 79–89. 10.3233/JAD-170498 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Gollan TH, Weissberger GH, Runnqvist E, Montoya RI, & Cera CM. (2012). Self-ratings of spoken language dominance: A Multilingual Naming Test (MINT) and preliminary norms for young and aging Spanish–English bilinguals. Bilingualism (Cambridge, England), 15(3), 594–615. 10.1017/S1366728911000332 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Grewal KS, O’Connell ME, Kirk A, MacDonald SWS, & Morgan D. (2023). Intraindividual variability measured with dispersion across diagnostic groups in a memory clinic sample. Applied Neuropsychology. Adult, 30(6), 639–648. 10.1080/23279095.2021.1970552 [DOI] [PubMed] [Google Scholar]
  21. Hair J, Black WC, Babin BJ, & Anderson RE (2010). Multivariate data analysis (7th ed.). Pearson Educational International. [Google Scholar]
  22. Hilborn JV, Strauss E, Hultsch DF, & Hunter MA (2009). Intraindividual variability across cognitive domains: Investigation of dispersion levels and performance profiles in older adults. Journal of Clinical and Experimental Neuropsychology, 31(4), 412–424. 10.1080/13803390802232659 [DOI] [PubMed] [Google Scholar]
  23. Hill BD, Rohling ML, Boettcher AC, & Meyers JE (2013). Cognitive intra-individual variability has a positive association with traumatic brain injury severity and suboptimal effort. Archives of Clinical Neuropsychology: The Official Journal of the National Academy of Neuropsychologists, 28(7), 640–648. 10.1093/arclin/act045 [DOI] [PubMed] [Google Scholar]
  24. Holmqvist SL, Thomas KR, Edmonds EC, Calcetas A, Edwards L, & Bangen KJ, Alzheimer’s Disease Neuroimaging Initiative. (2023). Cognitive dispersion is elevated in amyloid-positive older adults and associated with regional hypoperfusion. Journal of the International Neuropsychological Society: JINS, 29(7), 621–631. 10.1017/S1355617722000649 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Holtzer R, Verghese J, Wang C, Hall CB, & Lipton RB (2008). Within-person across-neuropsychological test variability and incident dementia. JAMA, 300(7), 823–830. 10.1001/jama.300.7.823 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Hopkins WG (2000). Measures of reliability in sports medicine and science. Sports Medicine, 30(1), 1–15. 10.2165/00007256-200030010-00001 [DOI] [PubMed] [Google Scholar]
  27. Ivanova I, Salmon DP, & Gollan TH (2013). The Multilingual Naming Test in Alzheimer’s Disease: Clues to the Origin of Naming Impairments. Journal of the International Neuropsychological Society: JINS, 19(3), 272–283. 10.1017/S1355617712001282 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Kälin AM, Pflüger M, Gietl AF, Riese F, Jäncke L, Nitsch RM, & Hock C. (2014). Intraindividual variability across cognitive tasks as a potential marker for prodromal Alzheimer’s disease. Frontiers in Aging Neuroscience, 6, 147. 10.3389/fnagi.2014.00147 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Karr JE, Garcia-Barrera MA, & Areshenkoff CN (2014). Executive functions and intraindividual variability following concussion. Journal of Clinical and Experimental Neuropsychology, 36(1), 15–31. 10.1080/13803395.2013.863833 [DOI] [PubMed] [Google Scholar]
  30. Kiselica AM, Karr JE, Mikula CM, Ranum RM, Benge JF, Medina LD, & Woods SP (2024). Recent advances in neuropsychological test interpretation for clinical practice. Neuropsychology Review, 34(2), 637–667. 10.1007/s11065-023-09596-1 [DOI] [PubMed] [Google Scholar]
  31. Kiselica AM, Kaser AN, Weitzner DS, Mikula CM, Boone A, Woods SP, Wolf TJ, & Webber TA (2024). Development and validity of norms for cognitive dispersion on the Uniform Data Set 3.0 neuropsychological battery. Archives of Clinical Neuropsychology: The Official Journal of the National Academy of Neuropsychologists, 39(6), 732–746. 10.1093/arclin/acae005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Kiselica AM, Webber TA, & Benge JF (2020). Using multivariate base rates of low scores to understand early cognitive declines on the uniform data set 3.0 Neuropsychological Battery. Neuropsychology, 34(6), 629–640. 10.1037/neu0000640 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Lee J, & Kim HJ (2022). Normal aging induces changes in the brain and neurodegeneration progress: Review of the structural, biochemical, metabolic, cellular, and molecular changes. Frontiers in Aging Neuroscience, 14, 931536. 10.3389/fnagi.2022.931536 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Lin SSH, & McDonough IM (2022). Intra-individual cognitive variability in neuropsychological assessment: A sign of neural network dysfunction. Neuropsychology, Development, and Cognition. Section B, Aging, Neuropsychology and Cognition, 29(3), 375–399. 10.1080/13825585.2021.2021134 [DOI] [PubMed] [Google Scholar]
  35. Malek-Ahmadi M, O’Connor K, Schofield S, Coon DW, & Zamrini E. (2018). Trajectory and variability characterization of the Montreal cognitive assessment in older adults. Aging Clinical and Experimental Research, 30(8), 993–998. 10.1007/s40520-017-0865-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Morgan EE, Woods SP, Rooney A, Perry W, Grant I, & Letendre SL, HIV Neurobehavioral Research Program HNRP Group. (2012). Intra-individual variability across neurocognitive domains in chronic hepatitis C infection: Elevated dispersion is associated with serostatus and unemployment risk. The Clinical Neuropsychologist, 26(4), 654–674. 10.1080/13854046.2012.680912 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Morris JC (1993). The Clinical Dementia Rating (CDR): current version and scoring rules. Neurology, 43(11), 2412–2414. 10.1212/wnl.43.11.2412-a [DOI] [PubMed] [Google Scholar]
  38. Morris JC, Weintraub S, Chui HC, Cummings J, Decarli C, Ferris S, Foster NL, Galasko D, graff-Radford N, Peskind ER, Beekly D, Ramos EM, & Kukull WA (2006). The Uniform Data Set (UDS): clinical and cognitive variables and descriptive data from Alzheimer Disease Centers. Alzheimer Disease and Associated Disorders, 20(4), 210–216. 10.1097/01.wad.0000213865.09806.92 [DOI] [PubMed] [Google Scholar]
  39. Partington JE, & Leiter RG (1949). Partington’s pathways test. Psychological Service Center Journal, 1, 11–20. [Google Scholar]
  40. Penheiro R, Webber TA, Kiselica AM, & Woods SP (2024). Executive functions are independently associated with cognitive dispersion in HIV disease. Archives of clinical neuropsychology: The official journal of the National Academy of Neuropsychologists, acae073. Advance online publication. 10.1093/arclin/acae073 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Portney LG, & Watkins MP (2000). Foundations of clinical research: Applications to practice. Prentice Hall. [Google Scholar]
  42. Possin KL, Tsoy E, & Windon CC (2021). Perils of race-based norms in cognitive testing: the case of former NFL players. JAMA Neurology, 78(4), 377–378. 10.1001/jamaneurol.2020.4763 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Sawilowsky SS (2009). New effect size rules of thumb. Journal of Modern Applied Statistical Methods, 8(2), 597–599. 10.22237/jmasm/1257035100 [DOI] [Google Scholar]
  44. Scott BM, Austin T, Royall DR, & Hilsabeck RC (2023). Cognitive intraindividual variability as a potential biomarker for early detection of cognitive and functional decline. Neuropsychology, 37(1), 52–63. 10.1037/neu0000867 [DOI] [PubMed] [Google Scholar]
  45. Stawski RS, MacDonald SWS, Brewster PWH, Munoz E, Cerino ES, & Halliday DWR (2019). A comprehensive comparison of quantifications of intraindividual variability in response times: A measurement burst approach. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences, 74(3), 397–408. 10.1093/geronb/gbx115 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, & Vandenbroucke JP, STROBE Initiative. (2007). The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. Annals of Internal Medicine, 147(8), 573–577. 10.7326/0003-4819-147-8-200710160-00010 [DOI] [PubMed] [Google Scholar]
  47. Webber TA, Kiselica AM, Mikula C, & Woods SP (2022). Dispersion-based cognitive intra-individual variability in dementia with Lewy bodies. Neuropsychology, 36(8), 719–729. 10.1037/neu0000856 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Webber TA, Lorkiewicz SA, Kiselica AM, & Woods SP (2024). Ecological validity of cognitive fluctuations in dementia with Lewy bodies. Journal of the International Neuropsychological Society: JINS, 30(1), 35–46. 10.1017/S1355617723000255 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Webber TA, Lorkiewicz S, Woods SP, Miller B, & Soble JR (2024). Does neuropsychological intraindividual variability index cognitive dysfunction, an invalid presentation, or both? Preliminary findings from a mixed clinical older adult veteran sample. Journal of Clinical and Experimental Neuropsychology, 46(6), 535–556. 10.1080/13803395.2024.2388096 [DOI] [PubMed] [Google Scholar]
  50. Weintraub S, Besser L, Dodge HH, Teylan M, Ferris S, Goldstein FC, Giordani B, Kramer J, Loewenstein D, Marson D, Mungas D, Salmon D, Welsh-Bohmer K, Zhou XH, Shirk SD, Atri A, Kukull WA, Phelps C, & Morris JC (2018). Version 3 of the Alzheimer Disease Centers’ Neuropsychological Test Battery in the Uniform Data Set (UDS). Alzheimer Disease and Associated Disorders, 32(1), 10–17. 10.1097/WAD.0000000000000223 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Woods SP, Scott JC, Dawson MS, Morgan EE, Carey CL, Heaton RK, & Grant I, HIV Neurobehavioral Research Center (HNRC) Group. (2005). Construct validity of Hopkins Verbal Learning Test-Revised component process measures in an HIV-1 sample. Archives of Clinical Neuropsychology: The Official Journal of the National Academy of Neuropsychologists, 20(8), 1061–1071. 10.1016/j.acn.2005.06.007 [DOI] [PubMed] [Google Scholar]
  52. Yesavage JA (1988). Geriatric depression scale. Psychopharmacology Bulletin, 24(4), 709–711. [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental appendix
NIHMS2176073-supplement-Supplemental_appendix.docx (23KB, docx)

Data Availability Statement

The data that support the findings of this study are available by request through the National Alzheimer’s Coordinating Center (NACC) at https://naccdata.org/requesting-data/data-request-process.

RESOURCES