Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Sep 1.
Published in final edited form as: Neuropsychology. 2022 May 23;37(6):628–635. doi: 10.1037/neu0000821

Cross-Sectional Associations of CSF Tau Levels With Rey’s AVLT: A Recency Ratio Study

Davide Bruno 1, Ainara Jauregi Zinkunegi 1, Nunzio Pomara 2,3, Henrik Zetterberg 4,5,6,7,8, Kaj Blennow 4,5, Rebecca Langhough Koscik 9,10, Cynthia Carlsson 9,10,11,12, Barbara Bendlin 10,11, Ozioma Okonkwo 10,11, Bruce P Hermann 9,13, Sterling C Johnson 9,10,11,12, Kimberly D Mueller 9,10,14
PMCID: PMC9681933  NIHMSID: NIHMS1838273  PMID: 35604714

Abstract

Objective:

The preeminent in vivo cerebrospinal fluid (CSF) biomarkers of Alzheimer’s disease (AD) are amyloid β 1–42 (Aβ42), phosphorylated Tau (p-tau), and total Tau (t-tau). The goal of this study was to examine how well traditional (total and delayed recall) and process-based (recency ratio [Rr]) measures derived from Rey’s Auditory Verbal Learning test (AVLT) were associated with these biomarkers.

Method:

Data from 235 participants (Mage = 65.5, SD = 6.9), who ranged from cognitively unimpaired to mild cognitive impairment, and for whom CSF values were available, were extracted from the Wisconsin Registry for Alzheimer’s Prevention. Bayesian regression analyses were carried out using CSF scores as outcomes, AVLT scores as predictors, and controlling for demographic data and diagnosis.

Results:

We found moderate evidence that Rr was associated with both CSF p-tau (Bayesian factor [BFM] = 5.55) and t-tau (BFM = 7.28), above and beyond the control variables, while it did not correlate with CSF Aβ42 levels. In contrast, total and delayed recall scores were not linked with any of the AD biomarkers, in separate analyses. When comparing all memory predictors in a single regression, Rr remained the strongest predictor of CSF t-tau levels (BFM = 3.57).

Conclusions:

Our findings suggest that Rr may be a better cognitive measure than commonly used AVLT scores to assess CSF levels of p-tau and t-tau in nondemented individuals.

Keywords: A/T/N biomarkers, Alzheimer’s disease, CSF tau, CSF Aβ42, recency ratio

In recent years, there have been calls for an unbiased approach to the diagnosis of Alzheimer’s disease (AD), leading to the A/T/N classification (Jack et al., 2018). According to this classification, the preeminent in vivo biomarkers of AD reflect amyloid β deposition (A), presence of pathological tau (T), and neurodegeneration (N), respectively (Jack et al., 2016, 2018). When using cerebrospinal fluid (CSF) measures, A will correspond to levels of amyloid β 1–42 (hereinafter, Aβ42), T to phosphorylated Tau (p-tau) levels, and N to levels of total Tau (t-tau).

Alongside biomarkers, however, there is also the need for developing sensitive cognitive measures that reflect the disease associated pathology indexed by biomarker levels (Bock et al., 2021). There are at least two good reasons to do this (Florean et al., 2022). First, cognitive tests are relatively affordable and can be easily available, including in areas where access to laboratory equipment is limited. Relatedly, cognitive tests are relatively unintrusive, thus facilitating data collection on participants who are averse to medical procedures. Second, because subtle neuropsychological changes can be observed already when individuals with elevated levels of AD biomarkers are still asymptomatic for the disease (e.g., Bruno, Mueller, et al., 2021; Mueller et al., 2020).

Among the available neuropsychological tests used for assessment of episodic memory, the Rey’s Auditory Verbal Learning test (hereinafter, AVLT; Rey, 1958) has been commonly used in older populations. In this test, subjects are read a list of 15 unrelated nouns five times and are asked to free recall these words after each presentation. Then a new 15-word list is tested (interference), followed again by free recall of the originally presented list. Finally, after about 20–30 min, subjects are asked to free recall the original list once again, ending with a recognition test. Most commonly, to evaluate free recall ability, total recall is calculated by summing the numbers of correctly recalled items across all five initial (learning) trials. Additionally, delayed recall is measured by the number of words recalled correctly after the 20- to 30-min delay.

An alternative approach to traditional scoring of neuropsychological tests is the examination of process scores (Kaplan, 1990). Process scores make assumptions on the underlying neurocognitive processes leading to test performance and aim to determine how and why an individual performed the way they did. Process scores have been shown to identify asymptomatic individuals at risk of subsequent cognitive decline (Bruno et al., 2013; Gicas et al., 2020; Talamonti et al., 2020; Thomas et al., 2018, 2020). An example of a free recall process score that can be extracted from AVLT data is the recency ratio (Rr; Bruno et al., 2016). Rr leverages the observation that individuals with AD present with good immediate memory recall for items learned at the end of a list (recency items, e.g., Foldi et al., 2003), while displaying very poor recency performance after a delay (Carlesimo et al., 1995). This ratio between immediate and delayed recency performance has been used to track, among other things, the level of risk in progressing to mild cognitive impairment (MCI) from a healthy baseline (Bruno et al., 2018; Egeland, 2021) and to AD from MCI (Turchetta et al., 2020). Rr has been shown to correlate also with CSF neurogranin, a biomarker of postsynaptic dysfunction (Bruno, Mueller, et al., 2021; Bruno, Reichert Plaska, et al., 2021), and to aid the differential diagnosis of AD (Turchetta et al., 2018). Higher Rr scores indicate more recency forgetting and, consequently, more overall risk of cognitive impairment. Bruno et al. (2018) showed that 82% of examined individuals who were cognitively intact at baseline and had Rr scores of 1 or above eventually converted to a classification of early MCI.

The goal of this study was to examine how well traditional (total and delayed recall) and process-based (Rr) measures, derived from Rey’s AVLT, were associated with cross-sectional CSF A/T/N biomarkers levels. To do this, we examined data from the Wisconsin Registry for Alzheimer’s Prevention (WRAP), a population study of middle-aged individuals with a family history of AD, based at the University of Wisconsin–Madison (Johnson et al., 2017). Our goal was to mimic the conditions of a clinical examination as much as possible by selecting the smallest available time lapse between CSF and AVLT visits and by considering data which, outside of the CSF biomarkers, would be readily available to a clinician, such as age and sex of the individual—hence, we opted not to include genetic markers of AD (e.g., Apolipoprotein E [APOE] gene status), as these are not typically available in routine examinations.

Method

Participants

Data were extracted from WRAP, an ongoing longitudinal cohort study based at the University of Wisconsin–Madison, USA. For the present study, WRAP participants were selected based on having completed at least two assessment visits: One for cognitive screening and one for a lumbar puncture leading to CSF extraction. All participants completed the AVLT and were classified as cognitively unimpaired—stable (CUS), cognitively unimpaired—declining (CUD), or with MCI via a consensus conference diagnosis. To classify individuals based on their cognitive status, WRAP uses a two-tiered consensus conference approach. First, an algorithm that identifies cases where impairment may exist is applied, based on whether or not they meet one or more of the following criteria: (a) the participant performs 1.5 SDs below the mean on factor scores or individual measures of memory, executive function, language, working memory, or attention (Clark et al., 2016; Koscik et al., 2014; Langhough Koscik et al., 2021); (b) cognitive performance on one or more tests falls below values used in other studies as cut points for clinical MCI diagnoses (e.g., Wechsler Memory Scale-Revised [WMS-R] Logical Memory II, Wechsler, 1987: Story A score <9: AD Neuroimaging Initiative, Petersen et al., 2010); or (c) an abnormal informant report indicates subjective cognitive or functional decline. Second, a consensus diagnosis is determined by a team including physicians, clinical neuropsychologists, and clinical nurse practitioners, based on cognitive, medical history, lifestyle, subjective cognitive complaints, and informant data, for each visit (Langhough Koscik et al., 2021). The CUD label is assigned when participants are performing lower than expected on internal norms, and the consensus review team has ruled out other causes, including worse diagnoses such as MCI or dementia. The MCI diagnosis follows the core clinical criteria (excluding biomarkers) from Albert et al. (2011; but see also Winblad et al., 2004), adopted by the National Institute on Aging (NIA)–Alzheimer’s Association. The core clinical criteria from McKhann et al. (2011) were used for dementia, without reference to biomarkers. From the total pool of 1,551 volunteers, 242 participants had CSF data from at least one visit, and 235 participants fulfilled all of the above inclusion criteria (see Table 1, for demographic information). Of these, 200 were CUS, 28 were CUD, and seven had MCI. All activities for this study were approved by the institutional review board of the University of Wisconsin–Madison and completed in accordance with the Helsinki Declaration. All participants provided informed consent prior to testing.

Table 1.

Demographics, CSF Measures, and Memory Tests Scores (M and SD) for the Study Participants

Characteristic Total CUS CUD MCI
N 235 200 28 7
Sex (females)    156 (66%)    135 (68%)   17 (61%)    4 (57%)
Age at LP (years)   65.5 (6.9)   65.1 (6.9)   67.2 (6.6)  70.4 (5.3)
Time elapsed (years)  1.4 (1.2)  1.5 (1.2)  1.2 (1.1)    1.9 (1.1)
WRAT-3 raw score   51.4 (4.6)   51.2 (4.7)   52.4 (3.4)  50.6 (4.6)
CSF Aβ42 (ng/L) 909.9 (405.9) 905.3 (393.0) 891.3 (484.5) 1115.8 (438.7)
CSF p-tau (ng/L)   18.9 (7.4)   18.9 (7.0)   17.8 (7.7)  26.2 (10.8)
CSF t-tau (ng/L) 214.1 (75.7) 213.3 (73.1) 198.7 (72.4)   299.3 (115.1)
Rr  1.3 (0.8)  1.3 (0.7)  1.5 (0.8)    2.9 (1.5)
AVLT total recall   51.9 (8.4)   53.2 (7.9)   45.5 (6.6)  40.6 (9.9)
AVLT delayed recall   10.7 (3.1)   11.2 (2.7)  8.5 (2.6)    4.3 (4.8)
Open in a new tab

Note. Time elapsed was calculated as an absolute value. CSF = cerebrospinal fluid; CUS = cognitively unimpaired—stable; CUD = cognitively unimpaired—declining; MCI = mild cognitive impairment; N = sample size; LP = lumbar puncture; WRAT-3 = Wide Range Achievement Test-3; Aβ42 = amyloid β 1–42; Rr = recency ratio; AVLT = Auditory Verbal Learning Test.

Procedure

WRAP procedures have been described previously (e.g., Johnson et al., 2017; Sager et al., 2005), but, briefly, WRAP is an ongoing longitudinal study based in Madison, WI, USA, of middle-aged individuals, who attend regular visits, typically every 2 years. Each participant completed self-report questionnaires on demographics, health history and lifestyle, in addition to clinical assessments, and a neuropsychological test battery. Some participants also underwent a lumbar puncture. The neuropsychological test battery included the Rey’s AVLT, described above.

CSF Collection

CSF was extracted using a Sprotte 24- or 25-gauge spinal needle, under fasting conditions. During each lumbar puncture visit, 22 mL of CSF was extracted, which was then combined, mixed, centrifuged, and aliquoted into tubes of 1.5 mL capacity. These tubes were stored within 30 min at −80 °C (for more details on the CSF procedure, see Van Hulle et al., 2021).

Biomarker Measurements

All CSF samples were assayed at the Clinical Neurochemistry Laboratory, University of Gothenburg, using the same batch of reagents under strict quality control procedures. Elecsys β-amyloid (1–42) CSF, Elecsys Phospho-Tau (181P) CSF, and Elecsys Total-Tau CSF were performed on a cobas e 601 analyzer, as previously described (Van Hulle et al., 2021).

Data Analysis Plan

To fulfill our study aims, we first carried out Bayesian linear regression analyses to test models with AVLT scores against null models including control variables (specified below). Bayesian analyses allow for the estimation of model plausibility, which permits comparison of models with different combinations of predictors, and for the determination of effect sizes with credible intervals (e.g., Teipel et al., 2021). For all analyses, the model prior was set to uniform, where all models are a priori equally likely, and the prior on parameters was set to the default Jeffreys–Zellner–Siow (JZS) prior probability distribution, which allows the Bayes factor to be the same regardless of unit of measurement. Credible intervals were set to 89%, which is considered more stable than 95% (Kruschke, 2014). To address potential issues with nonnormally distributed residuals in the regressions, Markov chain Monte Carlo (MCMC) sampling to each analysis was applied 1,000 times. The outcome variables were CSF levels of Aβ42, p-tau, and t-tau, in separate analyses. Control variables (forming the null models) were age at the lumbar puncture, time elapsed between lumbar puncture and memory assessment, sex, consensus diagnosis at lumbar puncture, and the Wide Range Achievement Test-3 (WRAT-3) reading subtest (raw score) at the baseline visit as a culturally reduced measure of cognitive reserve, within the North American context (Manly et al., 2002). Predictor variables were total recall, delayed recall, and Rr: We first looked at these predictors in separate analyses and then compared them directly in a single regression. Total recall is calculated by adding the numbers of correctly free recalled words across all five learning trials. Delayed recall is the total number of free recalled items at the delayed recall trial. Finally, Rr is calculated by dividing the number of correctly recalled recency words (i.e., the last four words presented) from the learning trial (Trial 1 of AVLT) by the corresponding number of correctly recalled recency words in the delayed recall trial. A correction also was applied, (immediate recency score + 1)/(delayed recency score + 1), to avoid missing data due to zero scores, as in Bruno et al. (2018). Analyses were conducted using Jeffreys’s Amazing Statistics Program; JASP (0.14; https://jasp-stats.org/).

Transparency and Openness

Data can be requested here: https://wrap.wisc.edu/data-requests/.

Results

Rr ranged from 0.20 to 5 in CUS, from 0.40 to 4 in CUD, and from 1.25 to 5 in MCI. AVLT total recall ranged from 28 to 71 in CUS, from 30 to 58 in CUD, and from 28 to 59 in MCI. AVLT delayed recall ranged from 4 to 15 in CUS, from 4 to 13 in CUD, and from 0 to 13 in MCI.

To test the hypothesis that Rr was associated with CSF levels of Aβ42, the null model was compared with the model including Rr. This analysis yielded no support to the hypothesis that Rr predicts Aβ42 in this sample, as the null model performed better than the model with Rr: The null model had a Bayesian factor (BFM) of 2.030, meaning that the odds in favor of the null model were about two times higher than the odds of the model with Rr. In contrast, the model with Rr had a BFM of 0.493. However, the Rr model (BFM = 5.546) outperformed the null model (BFM = 0.180) with p-tau as the outcome (i.e., the Rr model is over five times as likely as the null model), showing moderate evidence in favor of the Rr model, and suggesting that Rr predicts CSF p-tau levels in this sample better than the control variables. Rr had a posterior mean of 1.15, a SD of 0.71, and lower/higher 89% credible intervals of 0 and 2.053, respectively (note that credible intervals are determined by MCMC sampling). These values indicate that one added Rr point increases CSF p-tau levels by about 1.15 points, corresponding to ~6% of the mean CSF p-tau level detected in this sample. Analogously, the Rr model (BFM = 7.283) outperformed the null model (BFM = 0.137) with t-tau as the outcome, suggesting that Rr predicts CSF t-tau levels in this sample better than the control variables. Rr had a posterior mean of 12.882, SD = 7.223, and lower/higher 89% credible intervals of 0 and 21.502, respectively. These values indicate that one added Rr point increases CSF t-tau levels by 12.882 points, corresponding to roughly 6% of the mean t-tau level detected in this sample (see Figure 1). In contrast, models with total or delayed recall routinely underperformed the null model.

Figure 1. Plots of the Correlations Between Residuals Recency Ratio (X-Axis) and Residuals CSF Levels (Y-Axis).

Figure 1

Open in a new tab

Note. The residuals were calculated by partialing out all control variables (sex, age at LP, time elapsed, and WRAT-3 raw score). a. Plot of the correlation between recency ratio and CSF p-tau. b. Plot of the correlation between recency ratio and CSF t-tau. c. Plot of the correlation between recency ratio and CSF Aβ42. CSF = cerebrospinal fluid; Aβ42 = amyloid β 1–42; LP = lumbar puncture; WRAT-3 = Wide Range Achievement Test-3. See the online article for the color version of this figure.

When comparing models with Rr, total recall, and delayed recall directly in a single regression, we observed that the Rr model remained the best model with CSF t-tau as outcome. The Rr model achieved a BFM of 3.573, showing moderate evidence in favor of that model, with better odds than the second best model, which combined Rr and total recall, BFM = 2.549 (BF10 = 0.790, meaning that the observed data are 0.790 times as likely to occur under the model with Rr and total recall, as compared to the Rr-only model). A model with all memory scores also performed worse than the model with only Rr: BFM = 0.970, BF10 = 0.360. However, no model reached a threshold above anecdotal evidence (i.e., BFM = 3) with CSF p-tau levels, topping out at BFM = 2.786 when using Rr alone as a predictor. This result suggests that, while the model with Rr performed better than the null model and models with other memory predictors, the evidence is insufficient to draw firm conclusions. Finally, the null model performed best with CSF Aβ42 levels (BFM = 3.824), beating the model with Rr (BFM = 1.475, BF10 = 0.493), which was second best.

Discussion

The goal of this study was to examine how well traditional (total and delayed recall) and process-based (Rr) measures, derived from Rey’s AVLT, were associated with A/T/N biomarkers of AD in CSF: Aβ42, p-tau, and t-tau. Our results indicate that Rr outperformed both total and delayed AVLT recall measures. Rr was associated with both CSF p-tau and t-tau levels with moderate evidence, whereas there were no correlations between total and delayed recall and any of the A/T/N biomarkers. When combining all memory predictors in a single analysis, the odds in favor of the Rr model remained the highest when predicting CSF t-tau levels, while evidence in favor of the Rr model predicting CSF p-tau levels was only anecdotal.

High Rr scores are thought to depend on a reduction in long-term memory, due to a loss of consolidation ability, while reliance on phonological/echoic short-term memory remains relatively intact (Bruno et al., 2018; Turchetta et al., 2020). Consolidation ability is typically associated with function of the medial temporal lobe (Wixted, 2004; Wixted & Cai, 2013), an area that is implicated early in neurodegeneration and, particularly, in tauopathy (Maass et al., 2019; Tennant et al., 2021). Therefore, it is possible that Rr may be more sensitive than total and delayed AVLT recall scores for detecting development of tangle pathology and neurodegeneration in the medial temporal lobe. To note, unlike total and delayed recall, Rr takes into consideration loss of information from immediate to delayed recall, as it tracks a difference between performances, and hence may be more attuned to neurodegenerative damage. Consistent with this idea is the previous finding that Rr was sensitive to CSF neurogranin levels, a biomarker of postsynaptic dysfunction and possible neurodegeneration, in cognitively intact older individuals with major depression (Bruno, Mueller, et al., 2021; Bruno, Reichert Plaska, et al., 2021).

It should be noted that, upon discovering that Rr was associated with CSF p-tau levels, it was not surprising to find that Rr was also correlated with CSF t-tau levels. This is because CSF p-tau and t-tau levels are highly correlated in individuals who do not have dementia (e.g., Van Hulle et al., 2021). Indeed, in our own data, CSF p-tau and t-tau levels achieved a Pearson’s R value of 0.981, indicating extremely high association. Therefore, it would be informative to replicate these analyses in individuals presenting with more pronounced symptoms of a neurodegenerative disease.

The pattern of results observed in this study (i.e., Rr did not predict CSF Aβ42 levels but did predict CSF p-tau and t-tau levels) contrasts with that observed in a previous report (Bruno et al., 2019). In Bruno et al. (2019), we observed that Rr (but not total or delayed recall in the AVLT) was sensitive to CSF Aβ42 levels in people with MCI (n = 16) but was not sensitive to CSF tau levels in the same group. One difference across these reports is that the sample size of the MCI group in Bruno et al. (2019) was more than twice as large as that in the present study (n = 7). It is therefore conceivable that if we were to test a larger sample of people with MCI, all A/T/N biomarkers might be correlated with Rr. Another difference to note, albeit not large, lies in the average ages of the cognitively unimpaired group across articles: They were marginally younger (62.5, SD = 9.2) in the Bruno et al.’s article, where Rr was not found to predict AD CSF biomarkers levels, and older in the present data (65.5, SD = 6.9), although this difference is probably unlikely to have much of an impact on the results. Nevertheless, more research is needed to elucidate the predictive power of Rr compared to traditional AVLT cognitive scores, with regard to AD biomarkers.

Another AVLT serial position marker that has been used to identify asymptomatic individuals at risk for cognitive decline is primacy (i.e., memory for items at the beginning of a list; e.g., Gicas et al., 2020; La Rue et al., 2008), particularly when measured after a delay (Bruno et al., 2013; Talamonti et al., 2020). Therefore, we also wanted to consider, post hoc, whether delayed primacy had any impact on CSF levels of AD biomarkers in this cohort. We defined primacy as the first four items on the list, as per Bruno et al. (2013), and focused on performance in the delayed trial. We then ran the same regression analyses as above but using delayed primacy as the sole predictor (while maintaining the same covariates). None of the delayed primacy models reached the BFM threshold of “3,” and the best result was an anecdotal effect of delayed primacy on p-tau (BFM = 2.072). We can therefore conclude that, within these data, Rr outperformed delayed primacy when predicting CSF levels of AD biomarkers.

A potential limitation of the present article is that the sample constituted primarily White participants. Evidence from previous studies indicates that different racial and ethnic groups show differences in brain morphology, such as hippocampal volume (DeCarli et al., 2008), white matter hyperintensity volume (Brickman et al., 2008; Divers et al., 2013), and total cerebral brain volume (Stavitsky et al., 2010). Thus, current findings on Rr as a marker of A/T/N biomarkers need to be examined in diverse groups of research participants, and outside of North America, using culturally appropriate versions of the same tests. Another limitation, partly noted above, is the small number of individuals with MCI. Small samples are more susceptible to measurement noise, which may also explain the odd observation whereby CSF Aβ42 are higher rather than lower in the MCI groups compared to the other groups. Future tests should address this issue and be carried out with a larger sample. Finally, another possible limitation, which would also probably benefit from a larger sample size, is the fact that delayed recall scores (see Table 1) are much lower in the MCI group compared to the CUS group. This occurrence might make delayed recall less reliable as a predictor.

To summarize, the current results showed that Rr, the ratio between AVLT immediate and delayed performance scores at the recency position is a sensitive measure of CSF levels of p-tau and t-tau. Higher Rr scores, showing disproportionate loss of recency recall from immediate to delayed testing, are associated with an increase in CSF p-tau and t-tau levels, but not with CSF Aβ42 levels, when controlling for demographics and diagnosis. In contrast, either the AVLT total or delayed recall scores significantly predicted CSF levels of AD biomarkers. We wish to draw two conclusions. First, as argued previously (Bruno et al., 2016, 2018, 2019; Bruno, Mueller, et al., 2021; Bruno, Reichert Plaska, et al., 2021), we suggest that serial position values should be included in databases examining AD and other types of dementia. Second, we posit that Rr is a worthwhile measure to add to the clinician’s arsenal (see also Egeland, 2021) when evaluating individuals suspected to be on a trajectory toward neurodegeneration, and, as such, that it should be evaluated also as a possible addition to the MCI diagnostic criteria.

Key Points.

Question:

This article asks how well process-based measures derived from Rey’s Auditory Verbal Learning test (AVLT), such as the recency ratio (Rr), associate with cerebrospinal fluid (CSF) levels of Alzheimer’s disease (AD) biomarkers when compared to traditional measures (e.g., total and delayed recall).

Findings:

The results of this study indicate that Rr outperforms AVLT total and delayed recall in predicting CSF levels of t-tau cross sectionally.

Importance:

These findings highlight the importance of process-based measures for the identification of individuals at risk of neurodegeneration.

Next Steps:

More research is needed to examine Rr clinical value among individuals presenting with a neurodegenerative disease.

Acknowledgments

Wisconsin Registry for Alzheimer’s Prevention (WRAP) is funded by National Institutes of Health (NIH) Grant AG027161 with additional support for the cerebrospinal fluid (CSF) work by AG021155, AG062167, and AG037639. This secondary analysis of WRAP data was funded by a NIH–National Institute on Aging; NIA (R01 144 AAI8612) grant to Kimberly D. Mueller, in which Davide Bruno and Rebecca Langhough Koscik are coinvestigators.

Henrik Zetterberg is a Wallenberg scholar supported by grants from the Swedish Research Council (No. 2018-02532), the European Research Council (No. 681712), Swedish State Support for Clinical Research (No. ALFGBG-720931), the Alzheimer Drug Discovery Foundation (ADDF), USA (No. 201809-2016862), the Alzheimer’s Disease Strategic Fund and the Alzheimer’s Association (No. ADSF-21-831376-C, No. ADSF-21-831381-C, and No. ADSF-21-831377-C), the Olav Thon Foundation, the Erling-Persson Family Foundation, Stiftelsen för Gamla Tjänarinnor, Hjärnfonden, Sweden (No. FO2019-0228), the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement no. 860197 (MIRIADE), European Union Joint Program for Neurodegenerative Disorders (JPND2021-00694), and the U.K. Dementia Research Institute at University College London (UCL).

Davide Bruno played lead role in conceptualization, formal analysis, and writing of original draft; supporting role in funding acquisition; and equal role in methodology. Ainara Jauregi Zinkunegi played lead role in data curation; supporting role in writing of review and editing; and equal role in formal analysis. Nunzio Pomara played supporting role in conceptualization and writing of review and editing. Henrik Zetterberg played supporting role in data curation, funding acquisition, and methodology. Kaj Blennow played supporting role in data curation and funding acquisition. Rebecca Langhough Koscik played supporting role in formal analysis, funding acquisition, methodology, and writing of review and editing. Cynthia Carlsson played supporting role in funding acquisition. Barbara Bendlin played supporting role in funding acquisition and writing of review and editing. Ozioma Okonkwo played supporting role in funding acquisition and writing of review and editing. Bruce P. Hermann played supporting role in writing of review and editing. Sterling C. Johnson played supporting role in funding acquisition and writing of review and editing. Kimberly D. Mueller played lead role in funding acquisition and supporting role in writing of review and editing.

Footnotes

Data can be requested here: https://wrap.wisc.edu/data-requests/.

References

  1. Albert MS, DeKosky ST, Dickson D, Dubois B, Feldman HH, Fox NC, Anthony G, Holtzman DM, Jagust WJ, Petersen RC, Snyder PJ, Carrillo MC, Thies B, & Phelps CH (2011). The diagnosis of mild cognitive impairment due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimer’s & Dementia, 7(3), 270–279. 10.1016/j.jalz.2011.03.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bock JR, Russell J, Hara J, & Fortier D (2021). Optimizing cognitive assessment outcomes for Alzheimer’s disease by matching wordlist memory test features to scoring methodology. Frontiers in Digital Health, 3, Article 161. 10.3389/fdgth.2021.750549 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Brickman AM, Schupf N, Manly JJ, Luchsinger JA, Andrews H, Tang MX, Reitz C, Small SA, Mayeux R, DeCarli C, & Brown TR (2008). Brain morphology in older African Americans, Caribbean Hispanics, and Whites from Northern Manhattan. Archives of Neurology, 65(8), 1053–1061. 10.1001/archneur.65.8.1053 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bruno D, Gleason CE, Koscik RL, Pomara N, Zetterberg H, Blennow K, & Johnson SC (2019). The recency ratio is related to CSF amyloid beta 1–42 levels in MCI-AD. International Journal of Geriatric Psychiatry, 34(3), 415–419. 10.1002/gps.5029 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bruno D, Koscik RL, Woodard JL, Pomara N, & Johnson SC (2018). The recency ratio as predictor of early MCI. International Psychogeriatrics, 30(12), 1883–1888. 10.1017/S1041610218000467 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bruno D, Mueller KD, Betthauser T, Chin N, Engelman CD, Christian B, Koscik RL, & Johnson SC (2021). Serial position effects in the logical memory test: Loss of primacy predicts amyloid positivity. Journal of Neuropsychology, 15(3), 448–461. 10.1111/jnp.12235 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bruno D, Reichert C, & Pomara N (2016). The recency ratio as an index of cognitive performance and decline in elderly individuals. Journal of Clinical and Experimental Neuropsychology, 38(9), 967–973. 10.1080/13803395.2016.1179721 [DOI] [PubMed] [Google Scholar]
  8. Bruno D, Reichert Plaska C, Clark DPA, Zetterberg H, Blennow K, Verbeek MM, & Pomara N (2021). CSF α-synuclein correlates with CSF neurogranin in late-life depression. The International Journal of Neuroscience, 131(4), 357–361. 10.1080/00207454.2020.1744596 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Bruno D, Reiss PT, Petkova E, Sidtis JJ, & Pomara N (2013). Decreased recall of primacy words predicts cognitive decline. Archives of Clinical Neuropsychology, 28(2), 95–103. 10.1093/arclin/acs116 [DOI] [PubMed] [Google Scholar]
  10. Carlesimo GA, Sabbadini M, Fadda L, & Caltagirone C (1995). Different components in word-list forgetting of pure amnesics, degenerative demented and healthy subjects. Cortex, 31(4), 735–745. 10.1016/S0010-9452(13)80024-X [DOI] [PubMed] [Google Scholar]
  11. Clark LR, Koscik RL, Nicholas CR, Okonkwo OC, Engelman CD, Bratzke LC, Hogan KJ, Mueller KD, Bendlin BB, Carlsson CM, Asthana S, Sager MA, Hermann BP, & Johnson SC (2016). Mild cognitive impairment in late middle age in the Wisconsin registry for Alzheimer’s prevention study: Prevalence and characteristics using robust and standard neuropsychological normative data. Archives of Clinical Neuropsychology, 31(7), 675–688. 10.1093/arclin/acw024 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. DeCarli C, Reed BR, Jagust W, Martinez O, Ortega M, & Mungas D (2008). Brain behavior relationships among African Americans, Whites, and Hispanics. Alzheimer Disease and Associated Disorders, 22(4), 382–391. 10.1097/WAD.0b013e318185e7fe [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Divers J, Hugenschmidt C, Sink KM, Williamson JD, Ge Y, Smith SC, Bowden DW, Whitlow CT, Lyders E, Maldjian JA, & Freedman BI (2013). Cerebral white matter hyperintensity in African Americans and European Americans with type 2 diabetes. Journal of Stroke and Cerebrovascular Diseases, 22(7), e46–e52. 10.1016/j.jstrokecerebrovasdis.2012.03.019 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Egeland J (2021). Following H.N. over 21 years: Recency change and reduced retention predict later impairment in memory, and recency ratio may combine both effects. Neurocase, 27(2), 147–154. 10.1080/13554794.2021.1893339 [DOI] [PubMed] [Google Scholar]
  15. Florean I, Penolazzi B, Menichelli A, Pastore M, Cattaruzza T, Mazzon G, & Manganotti P (2022). Using the ATN system as a guide for the neuropsychological assessment of Alzheimer’s disease. Journal of Clinical and Experimental Neuropsychology, 43(9), 1–18. 10.1080/13803395.2022.2036327 [DOI] [PubMed] [Google Scholar]
  16. Foldi NS, Brickman AM, Schaefer LA, & Knutelska ME (2003). Distinct serial position profiles and neuropsychological measures differentiate late life depression from normal aging and Alzheimer’s disease. Psychiatry Research, 120(1), 71–84. 10.1016/S0165-1781(03)00163-X [DOI] [PubMed] [Google Scholar]
  17. Gicas KM, Honer WG, Wilson RS, Boyle PA, Leurgans SE, Schneider JA, & Bennett DA (2020). Association of serial position scores on memory tests and hippocampal-related neuropathologic outcomes. Neurology, 95(24), e3303–e3312. 10.1212/WNL.0000000000010952 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Jack CR Jr., Bennett DA, Blennow K, Carrillo MC, Dunn B, Haeberlein SB, Holtzman DM, Jagust W, Jessen F, Karlawish J, Liu E, Jose LM, Montine T, Phelps C, Rankin KP, Rowe CP, Scheltens E, Siemens E, Snyder HM, … the Contributors. (2018). NIA-AA research framework: Toward a biological definition of Alzheimer’s disease. Alzheimer’s & Dementia, 14(4), 535–562. 10.1016/j.jalz.2018.02.018 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Jack CR Jr., Bennett DA, Blennow K, Carrillo MC, Feldman HH, Frisoni GB, Hampel H, Jagust WJ, Johnson KA, Knopman DS, Petersen RC, Scheltens P, Sperling RA, & Dubois B (2016). A/T/N: An unbiased descriptive classification scheme for Alzheimer disease biomarkers. Neurology, 87(5), 539–547. 10.1212/WNL.0000000000002923 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Johnson SC, Koscik RL, Jonaitis EM, Clark LR, Mueller KD, Berman SE, Bendlin BB, Engelman CD, Okonkwo OC, Hogan KJ, Asthana S, Carlsson CM, Hermann BP, & Sager MA (2017). The Wisconsin registry for Alzheimer’s prevention: A review of findings and current directions. Alzheimer’s & Dementia: Diagnosis, Assessment & Disease Monitoring, 10, 130–142. 10.1016/j.dadm.2017.11.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kaplan E (1990). The process approach to neuropsychological assessment of psychiatric patients. Aphasiology, 2(1), 72–87. 10.1176/jnp.2.1.72 [DOI] [PubMed] [Google Scholar]
  22. Koscik RL, La Rue A, Jonaitis EM, Okonkwo OC, Johnson SC, Bendlin BB, Hermann BP, & Sager MA (2014). Emergence of mild cognitive impairment in late middle-aged adults in the Wisconsin registry for Alzheimer’s prevention. Dementia and Geriatric Cognitive Disorders, 38(1–2), 16–30. 10.1159/000355682 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kruschke J (2014). Doing Bayesian data analysis: A tutorial with R. JAGS, and Stan. Academic Press. [Google Scholar]
  24. La Rue A, Hermann B, Jones JE, Johnson S, Asthana S, & Sager MA (2008). Effect of parental family history of Alzheimer’s disease on serial position profiles. Alzheimer’s & Dementia, 4(4), 285–290. 10.1016/j.jalz.2008.03.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Langhough Koscik R, Hermann BP, Allison S, Clark LR, Jonaitis EM, Mueller KD, Betthauser TJ, Christian BT, Du L, Okonkwo O, Birdsill A, Chin N, Gleason C, & Johnson SC (2021). Validity evidence for the research category, “Cognitively Unimpaired—Declining,” as a risk marker for mild cognitive impairment and Alzheimer’s disease. Frontiers in Aging Neuroscience, 13, Article 688478. 10.3389/fnagi.2021.688478 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Maass A, Berron D, Harrison TM, Adams JN, La Joie R, Baker S, Mellinger T, Bell RK, Swinnerton K, Inglis B, Robinovici GD, Düzel E, & Jagust WJ (2019). Alzheimer’s pathology targets distinct memory networks in the ageing brain. Brain: A Journal of Neurology, 142(8), 2492–2509. 10.1093/brain/awz154 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Manly JJ, Jacobs DM, Touradji P, Small SA, & Stern Y (2002). Reading level attenuates differences in neuropsychological test performance between African American and White elders. Journal of the International Neuropsychological Society, 8(3), 341–348. 10.1017/S1355617702813157 [DOI] [PubMed] [Google Scholar]
  28. McKhann GM, Knopman DS, Chertkow H, Hyman BT, Jack CR Jr., Kawas CH, Klunk WE, Koroshetz WJ, Manly JJ, Mayeux R, Mohs RC, Morris RC, Rossor MN, Scheltens P, Carrillo MC, Thies B, Weintraub S, & Phelps CH (2011). The diagnosis of dementia due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimer’s & Dementia, 7(3), 263–269. 10.1016/j.jalz.2011.03.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Mueller KD, Koscik RL, Du L, Bruno D, Jonaitis EM, Koscik AZ, Christian BT, Betthauser TJ, Chin NA, Hermann BP, & Johnson SC (2020). Proper names from story recall are associated with beta-amyloid in cognitively unimpaired adults at risk for Alzheimer’s disease. Cortex, 131, 137–150. 10.1016/j.cortex.2020.07.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Petersen RC, Aisen PS, Beckett LA, Donohue MC, Gamst AC, Harvey DJ, Jack CR, Jagust WJ, Shaw LM, Toga AW, Trojanowski JQ, & Weiner MW (2010). Alzheimer’s disease neuroimaging initiative (ADNI): Clinical characterization. Neurology, 74(3), 201–209. 10.1212/WNL.0b013e3181cb3e25 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Rey A (1958). L’examen clinique en psychologie. Presses Universitaries De France. [Google Scholar]
  32. Sager MA, Hermann B, & La Rue A (2005). Middle-aged children of persons with Alzheimer’s disease: APOE genotypes and cognitive function in the Wisconsin registry for Alzheimer’s prevention. Journal of Geriatric Psychiatry and Neurology, 18(4), 245–249. 10.1177/0891988705281882 [DOI] [PubMed] [Google Scholar]
  33. Stavitsky K, Du Y, Seichepine D, Laudate TM, Beiser A, Seshadri S, DeCarli C, Wolf PA, & Au R (2010). White matter hyperintensity and cognitive functioning in the racial and ethnic minority cohort of the Framingham heart study. Neuroepidemiology, 35(2), 117–122. 10.1159/000313443 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Talamonti D, Koscik R, Johnson S, & Bruno D (2020). Predicting early mild cognitive impairment with free recall: The primacy of primacy. Archives of Clinical Neuropsychology, 35(2), 133–142. 10.1093/arclin/acz013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Teipel SJ, Dyrba M, Ballarini T, Brosseron F, Bruno D, Buerger K, Cosma N-C, Dechent L, Düzel E, Ewers M, Fliessbach K, Haynes J, Janowitz D, Kilimann I, Laske C, Maier C, Maier F, Metzger CD, Munk MH, & Heneka MT (2021). (Preprint). Association of cholinergic basal forebrain volume and functional connectivity with markers of inflammatory response in the Alzheimer’s disease spectrum. Journal of Alzheimer’s Disease, 85(3), 1267–1282. 10.3233/JAD-215196 [DOI] [PubMed] [Google Scholar]
  36. Tennant VR, Harrison TM, Adams JN, La Joie R, Winer JR, & Jagust WJ (2021). Fusiform Gyrus Phospho-Tau is associated with failure of proper name retrieval in aging. Annals of Neurology, 90(6), 988–993. 10.1002/ana.26237 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Thomas KR, Bangen KJ, Weigand AJ, Edmonds EC, Wong CG, Cooper S, Delano-Wood L, Bondi MW, & the Alzheimer’s Disease Neuroimaging Initiative. (2020). Objective subtle cognitive difficulties predict future amyloid accumulation and neurodegeneration. Neurology, 94(4), e397–e406. 10.1212/WNL.0000000000008838 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Thomas KR, Edmonds EC, Eppig J, Salmon DP, Bondi MW, & the Alzheimer’s Disease Neuroimaging Initiative. (2018). Using neuropsychological process scores to identify subtle cognitive decline and predict progression to mild cognitive impairment. Journal of Alzheimer’s Disease, 64(1), 195–204. 10.3233/JAD-180229 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Turchetta CS, De Simone MS, Perri R, Fadda L, Caruso G, De Tollis M, Caltagirone C, & Carlesimo GA (2020). Forgetting rates on the recency portion of a word list predict conversion from mild cognitive impairment to Alzheimer’s disease. Journal of Alzheimer’s Disease, 73(4), 1295–1304. 10.3233/JAD-190509 [DOI] [PubMed] [Google Scholar]
  40. Turchetta CS, Perri R, Fadda L, Caruso G, De Simone MS, Caltagirone C, & Carlesimo GA (2018). Forgetting rate on the recency portion of a word list differentiates mild to moderate Alzheimer’s disease from other forms of dementi. Journal of Alzheimer’s Disease, 66(2), 461–470. 10.3233/JAD-180690 [DOI] [PubMed] [Google Scholar]
  41. Van Hulle C, Jonaitis EM, Betthauser TJ, Batrla R, Wild N, Kollmorgen G, Andreasson U, Okonkwo O, Bendlin BB, Asthana S, Carlsson CM, Johnson SC, Zetterberg H, & Blennow K (2021). An examination of a novel multipanel of CSF biomarkers in the Alzheimer’s disease clinical and pathological continuum. Alzheimer’s & Dementia, 17(3), 431–445. 10.1002/alz.12204 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Wechsler D (1987). WMS-R: Wechsler memory scale-revised. Psychological Corporation. [Google Scholar]
  43. Winblad B, Palmer K, Kivipelto M, Jelic V, Fratiglioni L, Wahlund LO, Nordberg A, Bäckman L, Albert M, Almkvist O, Arai H, Basun H, Blennow K, De Leon M, DeCarli C, Erkinjuntti T, Giacobini E, Graff C, Hardy J, & Petersen RC (2004). Mild cognitive impairment—beyond controversies, towards a consensus: Report of the international working group on mild cognitive impairment. Journal of Internal Medicine, 256(3), 240–246. 10.1111/j.1365-2796.2004.01380.x [DOI] [PubMed] [Google Scholar]
  44. Wixted JT (2004). The psychology and neuroscience of forgetting. Annual Review of Psychology, 55, 235–269. 10.1146/annurev.psych.55.090902.141555 [DOI] [PubMed] [Google Scholar]
  45. Wixted JT, & Cai DJ (2013). Memory consolidation. The Oxford Handbook of Cognitive Neuroscience, 1, 436–456. 10.1093/oxfordhb/9780199988693.013.0021 [DOI] [Google Scholar]

RESOURCES