23289503, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/acn3.51245 by Readcube (Labtiva Inc.), Wiley Online Library on [20/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License BRIEF COMMUNICATION Accelerated long-term forgetting over three months in asymptomatic APOE ɛ4 carriers  Tort-Merino1,2 , Matti Laine3, Natalia Valech1, Jaume Olives1,2, Marıa Leo Adria  n 1, Mirian Ecay-Torres , Ainara Estanga , Pablo Martınez-Lage , Juan Fortea , 4 4 4 5 nchez-Valle1,2,6, Lorena Rami1,2,6,a & Antoni Rodrıguez-Fornells2,7,8,9,a Raquel Sa 1 Alzheimer’s Disease and other Cognitive Disorders Unit, Neurology Service, Hospital Clınic of Barcelona, Barcelona, Spain 2 Institute of Neurosciences, University of Barcelona, Barcelona, Spain 3 Department of Psychology,  Abo Akademi University, Turku, Finland 4 Centro de Investigacion y Terapias Avanzadas, Neurologıa, CITA-Alzheimer Foundation, San Sebasti an, Spain 5 Memory Unit, Department of Neurology, Hospital de la Santa Creu i Sant Pau and Institute of Biomedical Research, Barcelona, Spain 6 August Pi i Sunyer Biomedical Research Institute (IDIBAPS), Barcelona, Spain 7 Cognition and Brain Plasticity Group, Bellvitge Biomedical Research Institute-IDIBELL, Hospitalet de Llobregat, Spain 8 Department of Cognition, Development and Education Psychology, University of Barcelona, Hospitalet de Llobregat, Spain 9 Catalan Institution for Research and Advanced Studies (ICREA), Barcelona, Spain Correspondence Abstract Antoni Rodriguez-Fornells: Cognition and Brain Plasticity Group– IDIBELL, Campus de Accelerated long-term forgetting (ALF) refers to a rapid loss of information Bellvitge, L’hospitalet de Llobregat, 08097 over days or weeks despite normal acquisition/encoding. Notwithstanding its Barcelona, Spain. Tel: +34 934021038; Fax: potential relevance as a presymptomatic marker of cognitive dysfunction, no +34 934024268. E-mail: study has addressed the relationship between ALF and Alzheimer’s disease (AD) antoni.rodriguez@icrea.cat biomarkers. We examined ALF in APOE ɛ4 carriers versus noncarriers, and its relationships with AD cerebrospinal fluid (CSF) biomarkers. We found ALF Funding Information over three months in APOE ɛ4 carriers (F(1,19) = 5.60; P < 0.05; Cohen’s n y Ministerio de Ciencia, Innovacio d = 1.08), and this performance was associated with abnormal levels of the CSF Universidades Grant Number: PGC2018- Ab42/ptau ratio (r = .614; P < 0.01). Our findings indicate that ALF is detect- 099859-B-I00. European Regional able in at-risk individuals, and that there is a relationship between ALF and the Development Fund (ERDF). Carlos III Health pathophysiological processes underlying AD. Institute Grant Numbers: AC14/00014, PI043610, PI19/00745 Received: 15 September 2020; Revised: 21 October 2020; Accepted: 22 October 2020 Annals of Clinical and Translational Neurology 2021; 8(2): 477–484 doi: 10.1002/acn3.51245 a Co-senior authors. autosomal dominant disease mutation carriers1 and Introduction asymptomatic individuals with increased genetic risk Accelerated long-term forgetting (ALF) has been defined (APOE ɛ4 carriers)3 over a 1-week retention period. as a rapid loss of information over days or weeks despite Despite the potential relevance of ALF as a presymp- normal acquisition and encoding. Recently, the potential tomatic marker of subtle cognitive dysfunction, no study usefulness of ALF in the early detection of subtle cogni- has yet addressed the relationship between ALF and AD tive difficulties in asymptomatic stages of familial1 and biomarkers, or even evaluated ALF over periods beyond sporadic2,3 Alzheimer’s disease (AD) has attracted particu- one week. To fill this gap, we employed an effortful and lar attention. Two recent studies showed larger ALF in cognitively demanding associative memory task to explore ª 2020 The Authors. Annals of Clinical and Translational Neurology published by Wiley Periodicals LLC on behalf of American Neurological Association 477 This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. 23289503, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/acn3.51245 by Readcube (Labtiva Inc.), Wiley Online Library on [20/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License A. Tort-Merino et al. ALF during a 6-month follow-up in cognitively healthy rates) [e.g., 1-(one-week free recall score/ FLS), for one- carriers versus noncarriers of the APOE ɛ4 haplotype. week free forgetting rate; and 1-(one-week cued recall Moreover we examined the relationship between ALF and score/ CLS), for 1-week cued forgetting rate]. Recall rates AD cerebrospinal fluid (CSF) biomarker levels. The possi- at 3 months and 6 months were also compared with the ble findings derived from the present work would be of 1-week recall session. At the end of the 6-month session, a particular relevance for AD prevention trials, as well as verbal recognition task was given. Concerning the APOE for the assessment and monitoring of cognitively unim- carriers, one subject did not obtain the CLS and three sub- paired populations at risk of AD. jects refused to complete the 6-month assessment. In order to mathematically model the forgetting func- tions, we followed previous recommendations.7–9 Forget- Participants and methods ting curves are characterized by a curvilinear relation that The present participants represent a sub-sample of Tort- shows a rapid initial decline of information that is followed Merino et al.,2 and thus the methods description follows by a slower and longer decay, thus showing that informa- that paper. Among all the participants included in this pre- tion is lost in a larger extent after initial encoding. Previ- vious work, 11 subjects were APOE ɛ4 carriers (heterozy- ous studies on forgetting have shown that these functions gous for ɛ3 and ɛ4). In the present study, these 11 carriers (power and logarithmic) are the most accurate ones to were age-, sex-, and education-matched with 11 noncarriers describe forgetting curves. Information decay with time in (homozygous for ɛ3). The study was not pre-registered, and our sample was fitted better using a logarithmic function thus the present analyses are exploratory. All subjects under- [(y = a–bln(time)] when compared to a power function. went APOE genotyping, a lumbar puncture to determine Thus, each data point (4-time data points) for each subject CSF amyloid-ß (Ab42), total tau (tau), and phosphorylated and condition (free and cued recall conditions) was fitted tau (ptau) levels, a standard neuropsychological assessment using a nonlinear least-squares regression. Fit parameters to ensure that they were cognitively normal, and the cogni- were calculated based on the residual sum of squares and tively demanding Ancient Farming Equipment Test (AFE-T) showing the proportion of data variance accounted for to tap learning and long-term retention at 1 week, 3 and (R2). The slope (parameter b) and intercept (parameter a) 6 months. Participants were blind to APOE status and CSF were computed separately for each subject and condition. results. The mean time lapse between the lumbar puncture The slope captures the forgetting rate of encoded informa- and the AFE-T assessment was 2.09 (1.4) years. The neu- tion while the intercept represents the estimated initial level ropsychological battery encompassed four cognitive domains of performance (immediately after the last learning run). (memory, language, perception, and executive functions; see Demographical data, levels of CSF Aß42, CSF tau, and Tort-Merino et al.2 for tests details), and it included the CSF ptau were compared using Student t-tests for Cognitive Reserve Questionnaire (CRQ) for assessing cogni- independent samples and Chi-square analyses when tive reserve4 and the Subjective Cognitive Decline Question- appropriate. Analysis of variance (ANOVA) with post-hoc naire (SCD-Q)5 for measuring cognitive concerns. Bonferroni corrections and effect sizes expressed as The AFE-T calls for learning novel object/name pairs: it Cohen’s d were conducted to compare the between-group contains 24 black-and-white images of unfamiliar ancient scores on the AFE-T and the standard neuropsychological farming equipment taken from the AFE paradigm6 which tests. Magnitude scale for Cohen’s d has been suggested as: are paired with a pseudoword. The test consists of two ini- d = 0.2, small effect size, d = 0.5, medium effect size and tial learning sessions administered on two consecutive d = 0.8, large effect size.10 The AFE-T scores included in days. Each learning session include seven learning runs the analyses were the two final learning scores (i.e., FLS (range of performance score 0-24 points per run). The and CLS) and the forgetting measures (raw scores and for- AFE-T provides two final learning outcomes: (1) the free getting rates). Pearson’s bivariate correlations were used to learning score (FLS) and (2) the cued learning score assess the association between the forgetting measures (in- (CLS). Long-term recall is examined one week, three cluding the slope parameter from the modeled functions) months and six months after the initial learning phase, with AD CSF biomarkers and subjective cognitive ratings. including a visual recognition task, free recall, and cued recall (participants are not told in advance that they will Results be reassessed). Forgetting rates for free and cued recall scores at 1 week, 3 months, and 6 months were obtained Group differences between APOE ɛ4 carriers with comparisons to the initial learning phase. The forget- and noncarriers ting rate was defined as one minus the ratio between each delayed score (free or cued) and the score obtained on the We observed no group differences in age, sex, years of FLS (for free forgetting rates) or CLS (for cued forgetting education, or cognitive reserve. Regarding the AD 478 ª 2020 The Authors. Annals of Clinical and Translational Neurology published by Wiley Periodicals LLC on behalf of American Neurological Association 23289503, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/acn3.51245 by Readcube (Labtiva Inc.), Wiley Online Library on [20/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License A. Tort-Merino et al. Table 1. Demographics, CSF levels, and AFE-T results in the APOE ɛ4 carriers versus noncarriers Parameters APOE ɛ4 non-carriers APOE ɛ4 carriers F P1 Demographics Gender (% women) 72.7% 72.7% 0.002 1.00 Age 65.1 (6.5) [58-75] 65.8 (6.9) [56-77] 0.23 0.814 Years of education 11.0 (3.3) [8-18] 12.2 (4.3) [8-20] 0.78 0.446 CRQ 16.4 (3.9) [9-22] 15.4 (5.5) [6-22] 0.46 0.651 CSF levels Ab42 873.9 (294.1) pg/ml 470.5 (163.1) pg/ml 3.98 0.001** Tau 247.4 (98.7) pg/ml 307.1 (95.2) pg/ml 1.44 0.165 Ptau 53.8 (14.8) pg/ml 59.1 (15.7) pg/ml 0.81 0.423 AFE-T Free learning score (FLS) 17.8 (6.2) 14.2 (6.8) 1.62 0.218 Cued learning score (CLS) 21.0 (3.2) 18.6 (4.8) 1.87 0.187 1-week free recall 14.1 (6.5) 9.8 (7.4) 2.05 0.168 1-week cued recall 18.2 (4.6) 15.1 (5.9) 1.86 0.188 1-week FFR 0.15 (0.4) 0.41 (0.3) 2.54 0.127 1-week CFR 0.14 (0.1) 0.21 (0.2) 1.21 0.285 3-month free recall 7.1 (6.9) 3.4 (4.6) 2.18 0.155 3-month cued recall 13.3 (5.5) 7.6 (6.4) 4.99 0.037* 3-month FFR 0.59 (0.3) 0.82 (0.2) 3.76 0.067 3-month CFR 0.37 (0.2) 0.62 (0.3) 5.60 0.029* 6-month free recall 7.3 (6.2) 4.8 (5.4) 0.81 0.380 6-month cued recall 13.1 (5.9) 10.2 (7.2) 0.89 0.358 6-month FFR 0.58 (0.3) 0.74 (0.2) 1.61 0.222 6-month CFR 0.38 (0.2) 0.48 (0.3) 0.67 0.423 1-week to 3-month FFR 0.51 (0.3) 0.75 (0.2) 3.52 0.076 1-week to 3-month CFR 0.27 (0.2) 0.57 (0.3) 8.70 0.008** 1-week to 6-month FFR 0.49 (0.3) 0.70 (0.2) 2.50 0.132 1-week to 6-month CFR 0.27 (0.2) 0.46 (0.3) 2.14 0.161 1-week visual recognition 48.0 (0.0) 47.5 (1.5) 1.44 0.244 3-month visual recognition 46.3 (1.8) 45.7 (1.6) 0.71 0.409 6-month visual recognition 46.0 (1.5) 46.1 (0.8) 0.04 0.839 6-month verbal recognition 21.9 (2.3) 22.4 (1.6) 0.26 0.616 Data are presented as means (SD; standard deviation) [range]. Key: CRQ, cognitive reserve questionnaire; CSF, cerebrospinal fluid; Ab42, amyloid- b isoform 42; Tau, total tau; ptau, phosphorylated tau; FFR, free forgetting rate; CFR, cued forgetting rate. 1 p values were determined by Student’s t-test for demographics and CSF levels; 2 2 X statistic. *p < 0.05 **p < 0.01. biomarkers, there were no differences in CSF tau or p-tau (Figure 1B), the slope (forgetting rate parameter) was stee- levels, but CSF Ab42 levels were significantly lower (t per for the carriers (b = 9.1) compared to noncarriers (20) = 3.98; P < 0.01) in the APOE ɛ4 carrier group (b = 6.6). Importantly for the convergent validity of the (Table 1). There were no differences in any of the stan- different forgetting measures used (information decay dard neuropsychological tests. Neither did the AFE-T mathematical model and forgetting rates at each time point, exhibit group differences in the learning scores (Table 1). see results below), the 3-month cued forgetting rate and the cued slope were strongly correlated (r = .796; P < 0.01). Modeling of forgetting curves Forgetting rates at 1-week, 3- and 6-months Figures 1A,B show the forgetting curves as well as the decay- ing information function modeled for the whole group. We computed forgetting rates as in previous studies of Importantly, the fit of the logarithmic function to the group ALF1,3 (see Figure 1C,D). At 3 months, cued recall score data was nearly perfect in both groups, slightly better for the was significantly lower (F(1,19) = 4.99; P < 0.05; Cohen’s noncarriers (R2 = 99% of variance explained) compared to d = 0.99) in the carriers than in the noncarriers (Table 1; carriers (R2 = 97%). Particularly for cued recall Figure 1B). Besides, the APOE ɛ4 carriers evidenced ALF ª 2020 The Authors. Annals of Clinical and Translational Neurology published by Wiley Periodicals LLC on behalf of American Neurological Association 479 23289503, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/acn3.51245 by Readcube (Labtiva Inc.), Wiley Online Library on [20/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License A. Tort-Merino et al. Figure 1. Panel A shows the free recall scores and the free slopes for APOE ɛ4 carriers and noncarriers. Panel B shows the cued recall scores and the cued slopes. Panels C and D show the free and cued forgetting rates, respectively. Key: FLS, Free learning score; CLS, Cued learning score; 1W-3M, Forgetting rate between the 1-week and 3-month recall score; 1W-6M, Forgetting rate between 1-week and 6-month recall. Error bars represent 95% CIs. *P < 0.05. in their 3-month cued forgetting rate (F(1,19) = 5.60; Correlational analyses P < 0.05; Cohen’s d = 1.08), as well as in their cued for- getting rate between the 1-week and 3-month recall (F Correlational analyses in the whole sample showed a sig- (1,19) = 8.70; P < 0.01; Cohen’s d = 1.32; Post hoc nificant negative correlation between the 3-month cued power = 0.81) (Table 1; Figure 1D). No significant group forgetting rate and the ratio CSF Ab42/ptau (r = .614; differences were found on any of the free forgetting rates P < 0.01; Figure 2A). This correlation remained signifi- (Table 1; Figure 1C) or on the recognition scores cant after controlling for potential confounders such as (Table 1). age (r = .481; P < 0.05), years of education (r = .606; 480 ª 2020 The Authors. Annals of Clinical and Translational Neurology published by Wiley Periodicals LLC on behalf of American Neurological Association 23289503, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/acn3.51245 by Readcube (Labtiva Inc.), Wiley Online Library on [20/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License A. Tort-Merino et al. Figure 2. Panels A and B show the correlation between the ratio CSF Ab42/ phosphorylated tau and the 3-month cued forgetting rate and the cued slope from the AFE-T, respectively. Panels C and D show the correlations between the SCD-Q score and the 3-month cued forgetting rate and the cued slope, respectively. Key: Blue dots, APOE ɛ4 noncarriers; Red dots, APOE ɛ4 carriers; CSF, cerebrospinal fluid; Ab42, Amyloid-b isoform 42; ptau, CSF phosphorylated tau levels; SCD-Q, Subjective Cognitive Decline Questionnaire. P < 0.01) or the delay between CSF AD biomarkers (r = .636; P < 0.01; Figure 2C), and a significant negative acquisition and AFE-T testing (r = .548; P < 0.05). correlation was found between the cued slope and the Besides, a significant positive correlation between the cued SCD-Q (r = .508; P < 0.05; Figure 2D), showing that slope (from the modeled function in each individual) and higher forgetting was associated with higher cognitive the ratio Ab42/ptau (r = .458; P < 0.05; Figure 2B) was concerns. found, again indicating that higher forgetting was associ- ated to lower (more pathologic) levels of CSF AD Discussion biomarkers. A significant positive correlation was found between Using a new highly demanding associative memory test, the 3-month cued forgetting rate and the SCD-Q score AFE-T, we examined ALF in cognitively healthy APOE ɛ4 ª 2020 The Authors. Annals of Clinical and Translational Neurology published by Wiley Periodicals LLC on behalf of American Neurological Association 481 23289503, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/acn3.51245 by Readcube (Labtiva Inc.), Wiley Online Library on [20/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License A. Tort-Merino et al. carriers and its relationships with AD CSF biomarkers. Although the APOE e4 carriers also performed worse The results revealed ALF in cued recall over three months than noncarriers on the free recall, the lack of statistically in our well-characterized sample of APOE e3/e4 heterozy- significant between-group differences in these measures gotes. Moreover this performance was positively associ- might be related to task difficulty and the floor effects ated with core AD CSF biomarkers and subjective observed beyond the 1-week session. Also, the lack of sig- cognitive scores. nificant between-group differences in the 6-month session The effects of APOE e4 on memory performance in might be due to the missing data in the APOE carriers clinically normal adults and its relationship with amyloid group, which would also explain the slight improvement burden have been widely studied.11–13 As expected, our of this group compared to its performance at 3 months. results showed that APOE e4 carriers presented lower Importantly, both groups fitted the logarithmic func- CSF Ab levels than noncarriers (Table 1). However, data tion nearly perfect and particularly for cued recall, the on the link between ALF and AD is still limited. Recent APOE e4 carriers showed a steeper slope (comprehensive evidence from animal research showed accelerated forget- forgetting measure) compared to noncarriers, indicating ting with intact learning performance in a study using a an overall tendency to show larger information loss after model of familial AD (pre-pathological PDAPP mice).14 the initial learning in the carrier group, especially in In humans, ALF has been proposed as a potential cogni- between the first week and the 3-month period. It is tive marker for asymptomatic stages of AD. In an earlier important to note that the slope parameter from the report, our group provided evidence for subtle learning employed function strongly correlated with the AFE-T dysfunction and long-term forgetting in preclinical AD forgetting rate at 3 months, demonstrating that both subjects by employing the AFE-T.2 A more recent study measures are clearly associated and again pointing out to on a cohort of autosomal dominant AD families showed the idea that the information is lost to a larger extent one-week ALF in presymptomatic mutation carriers by during the initial period after encoding. These results may using three standard cognitive tests1. Moreover Zimmer- also explain why the correlations found are larger with man & Butler3 revealed ALF in asymptomatic APOE ɛ4 the 3-month forgetting rate and the slope. After three carriers by assessing 60 participants (20 homozygous for months the rate of information loss is very small and ɛ3, 20 heterozygous for ɛ3 and ɛ4, and 20 homozygous therefore it might be futile to evaluate ALF beyond this for ɛ4) with a standard memory test. Participants were time point. Taken together, our findings speak for the use asked to learn a 15-word list from the Rey Auditory Ver- of more sensitive measures with longer follow-up designs bal Learning Test (RAVLT)15 on four consecutive trials to when studying cognitively healthy at-risk subjects. ALF is an 80% accuracy criterion. In their analyses, APOE status emerging as a cognitive feature of presymptomatic AD, had no effects on memory encoding or short-term recall, highlighting the utility of long-term recall designs on but was associated with long-term forgetting over one monitoring cognitive changes in this population. week. Similar to the previous studies,1,3 we found a relation- In line with Zimmermann & Butler,3 we found intact ship between self-perceived subjective cognitive decline learning and ALF in APOE e4 carriers. However, here we and ALF. This is an important finding since subjective employed AFE-T in order to explore participants’ cogni- cognitive concerns have been suggested as a risk factor tion in a more comprehensive way. AFE-T requires form- for cognitive decline and dementia.17,18 Furthermore, the ing new associations or binding information without present results showed a strong correlation between ALF previous semantic knowledge, which is setting high (measured by cued forgetting rates and the slope parame- demands on cognitive processing,16 especially compared ter from the AFE-T) and the CSF Ab42/ptau ratio. Impor- to other standard episodic memory tasks. Regarding the tantly, CSF Ab42 and CSF ptau levels are valid indicators learning performance, AFE-T allowed a deeper analysis of of the abnormal protein deposits underlying AD patho- encoding processes by yielding free and cued recall scores. physiology (i.e., b-amyloid plaques and neurofibrillary Another strength is that we did not predefine any learn- tangles, respectively) and define AD as a specific neurode- ing accuracy criterion which limits scoring range and generative disease amongst other cognitive disorders.19 thereby the chances for finding between-group differences. Our finding on the correlation between ALF as measured The setup with AFE-T also allowed exploration of free by AFE-T and the ratio CSF Ab42/ptau provides further and cued recall for a longer time period, tracking ALF evidence for ALF as a promising candidate for a specific over 6 months after the learning phase. While the one- marker of AD-related subtle cognitive decline in presymp- week session appeared to be too close to initial learning, tomatic stages of the disease. Finally, it is important to the 3-month session was the optimal time-point to find note that given our relatively small sample size, validation ALF in our APOE e4 carriers, and cued recall measures of the present results is called for in future studies includ- showed higher sensitivity than spontaneous naming. ing larger samples and more robust power analyses. 482 ª 2020 The Authors. Annals of Clinical and Translational Neurology published by Wiley Periodicals LLC on behalf of American Neurological Association 23289503, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/acn3.51245 by Readcube (Labtiva Inc.), Wiley Online Library on [20/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License A. Tort-Merino et al. In sum, ALF is detectable in at-risk individuals, and novel information in preclinical Alzheimer’s disease stage there is a relationship between this cognitive measure 1. J Alzheimer’s Dis 2017;58(3):855–870. https://doi.org/10. and the pathophysiological processes underlying AD. 3233/JAD-161173. We strongly recommend the use of more demanding 3. Zimmermann JF, Butler CR. Accelerated long-term cognitive tests including long-term forgetting measures forgetting in asymptomatic APOE e4 carriers. Lancet for identifying and tracking the earliest cognitive man- Neurol 2018;17(5):394–395. https://doi.org/10.1016/S1474- ifestations in presymptomatic AD. The present results 4422(18)30078-4. show that accelerated forgetting is able to capture 4. Rami L, Valls-Pedret C, Bartres-Faz D, et al. Cognitive cognitive dysfunction earlier in the AD continuum. reserve questionnaire. Scores obtained in a healthy elderly Thus, they are particularly relevant for AD prevention population and in one with Alzheimer’s disease. Rev trials, in which regulatory agencies require a change Neurol 2011;52(4):195–201. https://doi.org/10.33588/rn. in cognition/functional status and not only a biomarker 5204.2010478. change. 5. Rami L, Mollica MA, Garcıa-Sanchez C, et al. The subjective cognitive decline questionnaire (SCD-Q): a validation study. J Alzheimer’s Dis 2014;41(2):453–466. Acknowledgments https://doi.org/10.3233/JAD-132027. 6. Laine M, Salmelin R. Neurocognition of new word This work has been supported by the Carlos III Health learning in the native tongue: lessons from the ancient Institute (projects PI043610 – AC14/00014 and PI19/ farming equipment paradigm. Lang Learn 2010;60 00745), integrated in the State Plan of Scientific and (2):25–44. https://doi.org/10.1111/j.1467-9922.2010. Technical Research and Innovation, and co-financed by 00599.x. the European Regional Development Fund (ERDF) “Una 7. Rubin DC, Wenzel AE. One hundred years of forgetting: a manera de hacer Europa”. This study has been funded by quantitative description of retention. Psychol Rev 1996;103 Ministerio de Ciencia, Innovaci on y Universidades (4):734–760. (MCIU), which is part of Agencia Estatal de Investigaci on 8. Wixted JT, Ebbesen EB. On the form of forgetting. (AEI), through the grant "Generaci on del Conocimiento", Psychol. Sci. 1991;2(6):409–415. number PGC2018-099859-B-I00. (Co-funded by Euro- 9. Wixted JT, Ebbesen EB. Genuine power curves in pean Regional Development Fund. ERDF, a way to build forgetting: a quantitative analysis of individual subject Europe). We thank CERCA Programme/Generalitat de forgetting functions. Mem Cogn 1997;25(5):731–739. Catalunya for institutional support. The authors thank all https://doi.org/10.3758/bf03211316. the volunteers for their participation in this study. 10. Cohen J. Statistical power analysis for the behavioral sciences. USA: Lawrence Erlbaum Associates, 1988. 11. Lim YY, Ellis KA, Ames D, et al. Ab amyloid, cognition, Author Contributions and APOE genotype in healthy older adults. Alzheimer’s A.R.F, L.R., M.L., and A.T.M. contributed to the concep- Dement 2013;9(5):538–545. https://doi.org/10.1016/j.jalz. tion and design of the study. All authors contributed to 2012.07.004. acquisition and analysis of the data. A.T.M., A.R.F., M.L., 12. Mormino EC, Betensky RA, Hedden T, et al. Amyloid and and L.R. contributed to drafting a significant portion of APOE e4 interact to influence short-term decline in the manuscript. All authors have reviewed the manuscript preclinical Alzheimer disease. Neurology 2014;82 and approved the final version. (20):1760–1767. https://doi.org/10.1212/WNL. 0000000000000431. 13. Kantarci K, Lowe V, Przybelski SA, et al. APOE modifies Conflicts of Interest the association between Ab load and cognition in cognitively normal older adults. Neurology 2012;78 Nothing to report. (4):232–240. https://doi.org/10.1212/WNL.0b013e 31824365ab. References 14. Beglopoulos V, Tulloch J, Roe AD, et al. Early detection of 1. Weston PSJ, Nicholas JM, Henley SMD, et al. Accelerated cryptic memory and glucose uptake deficits in pre- long-term forgetting in presymptomatic autosomal pathological APP mice. Nat Commun 2016;7:1–10. dominant Alzheimer’s disease: a cross-sectional study. https://doi.org/10.1038/ncomms11761. Lancet Neurol 2018;17(2):123–132. https://doi.org/10.1016/ 15. Rey A. L’examen clinique en psychologie. Paris: Presses S1474-4422(17)30434-9. Universitaire de France, 1964. 2. Tort-Merino A, Valech N, Pe~ naloza C, et al. Early 16. Rentz DM, Amariglio RE, Becker JA, et al. Face-name detection of learning difficulties when confronted with associative memory performance is related to ª 2020 The Authors. Annals of Clinical and Translational Neurology published by Wiley Periodicals LLC on behalf of American Neurological Association 483 23289503, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/acn3.51245 by Readcube (Labtiva Inc.), Wiley Online Library on [20/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License A. Tort-Merino et al. amyloid burden in normal elderly. Neuropsychologia 18. Reisberg B, Shulman MB, Torossian C, et al. Outcome over 2011;49(9):2776–2783. https://doi.org/10.1016/j.ne seven years of healthy adults with and without subjective uropsychologia.2011.06.006. cognitive impairment. Alzheimer’s Dement 2010;6(1):11–24. 17. Jessen F, Wiese B, Bachmann C, et al. Prediction of https://doi.org/10.1016/j.jalz.2009.10.002. dementia by subjective memory impairment effects of 19. Jack CR, Bennett DA, Blennow K, et al. NIA-AA Research severity and temporal association with cognitive Framework: Toward a biological definition of Alzheimer’s impairment. Arch Gen Psychiatry 2010;67(4):414–422. disease. Alzheimer’s Dement 2018;14(4):535–562. https://d https://doi.org/10.1001/archgenpsychiatry.2010.30. oi.org/10.1016/j.jalz.2018.02.018. 484 ª 2020 The Authors. Annals of Clinical and Translational Neurology published by Wiley Periodicals LLC on behalf of American Neurological Association