Archives of Clinical Neuropsychology 32 (2017) 142–154

Using Robust Normative Data to Investigate the Neuropsychology

Downloaded from https://academic.oup.com/acn/article/32/2/142/2655685 by Universidad de Malaga user on 14 November 2023

of Cognitive Aging
Karra D. Harrington1,2,*, Yen Ying Lim1,3, David Ames4,5, Jason Hassenstab6,7,8,
Stephanie Rainey-Smith9,10, Joanne Robertson1, Olivier Salvado11, Colin L. Masters1, Paul Maruff1,3,
for the AIBL Research Group12
1
The Florey Institute of Neuroscience and Mental Health, The University of Melbourne, Parkville, VIC, Australia
2
Cooperative Research Centre (CRC) for Mental Health, Carlton South, VIC, Australia
3
CogState Ltd., Melbourne, VIC, Australia
4
Academic Unit for Psychiatry of Old Age, Department of Psychiatry, The University of Melbourne, Parkville, VIC, Australia
5
National Ageing Research Institute, Parkville, VIC, Australia
6
Charles F. and Joanne Knight Alzheimer’s Disease Research Center, St. Louis, MO, USA
7
Department of Neurology, Washington University School of Medicine, St. Louis, MO, USA
8
Department of Psychology, Washington University, St. Louis, MO, USA
9
Centre of Excellence for Alzheimer’s Disease Research and Care, School of Exercise, Biomedical and Health Sciences,
Edith Cowan University, Perth, WA, Australia
10
Sir James McCusker Alzheimer’s Disease Research Unit, Hollywood Private Hospital, Nedlands, WA, Australia
11
CSIRO Preventative Health National Research Flagship, The Australian e-Health Research Centre—BioMedIA, Herston, QLD, Australia
12
https://aibl.csiro.au
*Corresponding author at: Florey Institute of Neuroscience and Mental Health, 155 Oak Street, Parkville, VIC 3052, Australia.
Tel.: 03-9388-1633; fax: 03-9387-5061.
E-mail address: karra.harrington@florey.edu.au
Editorial Decision 14 November 2016; Accepted 21 November 2016

Abstract
Objective: The extent to which increasing age is associated with impairment in cognitive function, termed cognitive aging, may have
been overestimated in prior studies. The inclusion of individuals with severe or uncontrolled systemic medical illness or prodromal neurode-
generative disease in normal aging samples is likely to bias estimates toward lower cognitive performance and inflate estimates of
variability.
Method: Unbiased estimates of cognitive aging in 658 adults aged 60–84, who underwent rigorous screening to ensure their general and
cognitive health, were computed. The first study screened the psychometric properties of a battery of neuropsychological tests in order to
identify those with optimal properties to evaluate cognitive aging. The second study used the selected tests to compare baseline performance
within 5-year age bands from 60 to 84.
Results: The first study identified a battery of 12 tests that provided reliable measures of memory, psychomotor speed, attention, and exec-
utive function and were appropriate for investigating age-related cognitive changes. The second study observed moderate to large age-
related impairment for performance on tests of complex psychomotor function, category fluency, verbal learning, and verbal and visual
memory. No, or only small, age effects were observed for working memory, phonemic fluency, learning of visual information, and reaction
time.
Conclusions: These data suggested that while increasing age is associated with impairment in cognitive function, this impairment is less
severe and is evident only on more complex neuropsychological tests than estimated previously in samples selected using less rigorous crite-
ria to ensure cognitive health.
Keywords: Elderly/geriatrics/aging; Assessment; Executive functions; Fluency (verbal/nonverbal); Learning and memory; Norms/normative studies

© The Author 2016. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
doi:10.1093/arclin/acw106 Advance Access publication on 8 December 2016
 K.D. Harrington et al. / Archives of Clinical Neuropsychology 32 (2017); 142–154 143

Introduction

Relationships between increasing age and decreasing sensory acuity, reductions in mobility and strength, and increasing
prevalence of systemic illness are well described (Rantakokko, Mänty, & Rantanen, 2013; Melis, Marengoni, Angleman, &
Fratiglioni, 2014; Heine & Browning, 2015). However, there remains uncertainty about the extent to which increasing age is
associated with a reduction in cognitive abilities independent of central nervous system (CNS) disease, termed cognitive
aging.
Studies of cognitive aging conclude that increasing age is associated with impairment across multiple cognitive domains

Downloaded from https://academic.oup.com/acn/article/32/2/142/2655685 by Universidad de Malaga user on 14 November 2023

(Deary et al., 2009; Lindenberger, 2014). Cross-sectional comparisons of older adults of different ages indicated that the
greatest age-related impairment (e.g., ~1–1.5 standard deviation [SD]) occurs for psychomotor speed and episodic memory
(Schaie, 2005; van Hooren et al., 2007; Salthouse, 2010; Bates & Wolbers, 2014; Bonsang & Dohmen, 2015). From a neuro-
psychological perspective, however, the criteria used to define the healthy aging samples used in many studies of cognitive
aging have lacked precision and this has limited the extent to which their data reflect true cognitive aging. For example,
healthy aging has been defined as the absence of a dementia diagnosis (e.g., Deary, Johnson & Starr, 2010; Brown et al.,
2012), normal scores on brief cognitive screening tools such as the Mini Mental State Examination (MMSE) or Montreal
Cognitive Assessment (e.g., Deary et al., 2010; Bates & Wobers, 2014), self-reports of good health on health questionnaires
(e.g., Bates & Wobers, 2014), or that the individual is community-dwelling (e.g., Gerstorf, Ram, Hoppmann, Willis, &
Schaie, 2011). Although such criteria will likely identify and exclude individuals with severe cognitive impairment or demen-
tia, they are unlikely to exclude those with subtle cognitive impairment or prodromal CNS disease (Deary et al., 2009;
Clouston, Glymour, & Terrera, 2015). Furthermore, increasing age is associated with increasing prevalence of systemic ill-
nesses, such as cardiovascular, endocrine, immune, or inflammatory disease, which themselves are also associated with subtle
impairments in cognition (Bergman, Blomberg, & Almkvist, 2007; Melis et al., 2014). While some aging studies have used
objective measures to determine the health status of their sample (e.g., Gerstorf et al. 2011 extract and code medical data),
others rely on self-reported medical history (e.g., Bates & Wobers, 2014) or do not report their methods for determining gen-
eral health status (e.g., Deary et al., 2010; Brown et al., 2012). Inclusion of adults with severe or uncontrolled systemic dis-
ease in samples used to study healthy aging will likely bias estimates of cognitive change negatively; particularly in older
groups where their prevalence increases (Bergman et al., 2007).
Neuropsychological estimates of healthy aging require more rigorous criteria be applied to exclude individuals with sub-
clinical CNS disease and with severe or uncontrolled systemic illness. For example, the development of robust normative data
is based on exclusion of older individuals with subclinical CNS disease through the use of longitudinal study where data for
individuals considered to be healthy initially, but who ultimately progress to meet clinical criteria for dementia, are removed
from estimates of normal performance (e.g., Sliwinski, Lipton, Buschke, & Stewart, 1996; De Santi et al., 2008; Clark et al.,
2016). Large prospective studies of neurodegenerative disease conducted within a medical context offer an opportunity to
apply the robust normative methodology to the development of neuropsychological models of cognitive aging. These studies
generally conduct thorough assessments of both physical and neurological health, often over multiple visits, in older indivi-
duals who are at risk of the disease of interest (e.g., Roberts et al., 2008; Petersen et al., 2010). Therefore, classification of
older individuals as being cognitively and physically healthy is based on the outcome of rigorous, objective, and repeated
assessment. This allows the identification and exclusion of individuals with severe or uncontrolled systemic illness as well as
those with subclinical CNS disease.
There is however a limitation in using neuropsychological data from prospective studies of neurodegenerative disease to
study the neuropsychology of aging, in that test batteries in such studies are designed primarily to identify cognitive dysfunc-
tion relevant to the disease of interest. For example, test batteries used to study the development of Alzheimer’s disease (AD)
focus on episodic memory whereas those used to study the development of Parkinson’s disease (PD) focus mainly on proces-
sing speed (Ellis et al., 2009; Weintraub et al., 2015). Such batteries may therefore not assess cognitive function in a breadth
sufficient to understand cognitive aging. They may also contain tests that do not have sufficient sensitivity to cognitive
changes in healthy older adults who typically perform at maximum levels (e.g., Agrell & Dehlin, 1998; Jefferson et al.,
2007). The resultant skewed data distributions reduce the reliability of performance scores and hence decrease the sensitivity
of tests to discriminate subtle differences in cognition (Storandt & Morris, 2010). Given these potential limitations, the psy-
chometric properties of data from neuropsychological test batteries used in prospective studies of neurodegenerative disease
should be evaluated carefully prior to their application to study cognitive aging.
The aim of this study was to estimate the effect of age on cognition by studying the performance on neuropsychological
tests of a large group of older adults whose physical and mental health had been assured from extensive repeated assessment
as part of their participation in a large prospective study of the development and evolution of AD. First, an analysis of the
psychometric properties of the neuropsychological test battery was conducted to identify neuropsychological tests optimal for
144 K.D. Harrington et al. / Archives of Clinical Neuropsychology 32 (2017); 142–154

understanding cognitive aging. These tests were then used to characterize the effect of age on cognition. The first hypothesis
was that older age would be associated with worse cognitive performance on all neuropsychological tests. The second hypoth-
esis was that performance on tests of psychomotor speed and episodic memory would show the greatest impairment with
increasing age.

Materials and Method

Participants

Downloaded from https://academic.oup.com/acn/article/32/2/142/2655685 by Universidad de Malaga user on 14 November 2023

Fig. 1 summarizes the sample screening process. All participants were aged between 60 and 85 and were selected from the
cognitively normal (CN) group of the Australian Imaging Biomarkers and Lifestyle (AIBL) study (Ellis et al., 2009). A con-
sensus panel of geriatricians, neurologists, psychiatrists, and neuropsychologists determined clinical classifications after re-
viewing participant assessment results. Participants were classified as CN if their MMSE was greater than 24 and if they did
not meet criteria for mild cognitive impairment (MCI; Winblad et al., 2004) or dementia (McKhann et al., 1984). The AIBL
study was approved by and complied with the regulations of three institutional research and ethics committees (St Vincent’s
Health, Austin Health, and Edith Cowan University). All participants provided written informed consent prior to participating
in the study.
The AIBL study initially recruited participants via a media appeal, as well as via referrals from local physicians (for full
details, see Ellis et al., 2009). Prior to enrollment, individuals were screened for significant medical and psychiatric history
with a telephone interview. At this stage, individuals were excluded for a history of non-AD dementia, Parkinson’s disease,
schizophrenia, bipolar disorder, cancer (other than basal skin carcinoma) within the last 2 years, uncontrolled diabetes, current
depression (indicated by a Geriatric Depression Scale score of >5), or current regular alcohol consumption exceeding two
standard drinks per day for women or four per day for men. Individuals who satisfied criteria on the phone screen then at-
tended the research unit to undergo detailed cognitive and neurological assessment (described in Ellis et al., 2009).
At the research center, all individuals underwent cognitive and neurological assessment, provided a detailed medical his-
tory, completed assessment of vital signs (blood pressure, heart rate, body mass index [BMI]), and provided a blood sample.
The study team also reviewed blood pathology results for all participants to identify any results indicative of systemic illness.

Fig. 1. Study sample screening process.

 K.D. Harrington et al. / Archives of Clinical Neuropsychology 32 (2017); 142–154 145

The classification scheme of Bergman and colleagues (2007) was applied to these data. Individuals classified as having
significant clinical CNS or systemic illness were excluded from this study. To ensure the absence of progressive CNS
disease, neuropsychological and clinical data from the follow-up (up to 72 months) assessments were also reviewed.
Neuropsychological data for individuals who had not completed at least one follow-up assessment, who progressed to meet
clinical criteria for MCI or dementia, or who later withdrew from the AIBL study for reasons of ill health (e.g., cancer or
stroke) were excluded from the current analyses.

Measures

Downloaded from https://academic.oup.com/acn/article/32/2/142/2655685 by Universidad de Malaga user on 14 November 2023

Demographic and clinical characteristics. Age, gender, and education were self-reported by participants. Trained research as-
sistants recorded medical history, assessed depression and anxiety symptoms using the Hospital Anxiety and Depression
Scale (HADS; Zigmond & Snaith, 1983), and administered the Wechsler Test of Adult Reading (WTAR; Wechsler, 2001) to
estimate IQ.

Neuropsychological tests. The AIBL neuropsychological test battery assesses across the cognitive domains of executive func-
tion, episodic memory (immediate and delayed recall, and recognition), language, psychomotor speed and attention, and
visuospatial functioning. The battery consists of 14 tests, including a combination of traditional paper and pencil neuropsy-
chological tasks and computerized tasks from the CogState Brief Battery (CBB). Briefly, the CBB consisted of four tasks that
require the participant to respond to playing card stimuli presented on the computer screen. These included the Detection
(reaction time), Identification (choice reaction time), One Card Learning (visual recognition memory), and One Back (working
memory) tasks (Lim et al., 2012). Responses are made by pressing an external response button as quickly as possible follow-
ing stimuli presentation and are in a Yes/No format. The complete neuropsychological battery, including the CBB, and
administration procedures have been described in detail previously (Ellis et al., 2009; Lim et al., 2012). Although many of
these tests require multiple areas of cognition for optimal performance, each has been classified according to the main cogni-
tive domain assessed (Lezak, 2012).
Executive Function was assessed by Digit Span subtest total score (forward + backward) of the Wechsler Adult
Intelligence Scale-Third edition (WAIS-III; Wechsler, 1997), D-KEFS Verbal Fluency including Letter Fluency (FAS),
Category Fluency (Animals Names; Fruit Furniture Total), and Category Switching (Fruit Furniture Switching; Delis, Kaplan,
& Kramer, 2001), the Stroop task Interference score (Victoria version; Strauss, Sherman, & Spreen, 2006), and the One Back
task from the CBB (Lim et al., 2012); Episodic memory was assessed by the California Verbal Learning Test-Second edition
(CVLT-II; Delis, Kramer, Kaplan, & Ober, 2000), Logical Memory I and II Story A from the Wechsler Memory Scale-
Revised (Wechsler, 1987), the Rey Complex Figure Test (RCFT; Meyers & Meyers, 1995), and the One Card Learning task
from the CBB (Lim et al., 2012). Language was assessed by the 30-item Boston Naming Test (BNT; Saxton et al., 2000);
Psychomotor Speed and Attention were assessed by the Digit Symbol Coding subtest of the WAIS-III (Wechsler, 1997),
the Stroop task Dots, Words, and Colors time scores (Strauss et al., 2006), and the Detection and Identification tasks from the
CBB (Lim et al., 2012); and Visuospatial Function was assessed by the Clock Drawing task (Strauss et al., 2006) and the
Copy trial of the RCFT (Meyers & Meyers, 1995).

Procedure

Participants fasted overnight prior to attending the research facility for clinical assessment. Written informed consent was
first obtained, followed by collection of an 80-mL blood sample (used for genotyping and full blood analyses for medical
assessment) and provision of breakfast. Participants then completed self-report questionnaires including demographic details
and the HADS. Trained research assistants conducted a clinical interview and the neuropsychological assessment according to
standardized protocols in a single session of approximately 120 min, followed by a medical examination including measure-
ment of height, weight, blood pressure, heart rate, and abdominal circumference.

Study 1: Psychometric Characteristics of Neuropsychological Tests of Aging

The aim of the first study was to identify neuropsychological tests from the battery that produced optimal performance
data to study age effects on cognition in healthy older adults.
146 K.D. Harrington et al. / Archives of Clinical Neuropsychology 32 (2017); 142–154

Data Analysis

Data frequency distributions, skewness values, missing data cases and test-retest reliability over 18 months were gener-
ated for each neuropsychological test outcome measure. Where a neuropsychological test had multiple outcome measures
each of these was examined independently. The process and results of data screening are summarized in Table 1. Data fre-
quency distributions were visually inspected to identify individual data points that were outliers. These data points were
subsequently excluded from analyses for that measure. As there are no guidelines or recommendations for the classifica-
tion of ceiling or floor effects in neuropsychological contexts, in the present study, a conservative criterion was defined
where ceiling effects were classified to have occurred if ≥20% of the data were at the maximum possible score. Similarly,

Downloaded from https://academic.oup.com/acn/article/32/2/142/2655685 by Universidad de Malaga user on 14 November 2023

floor effects were determined to have occurred if ≥20% of the data were at the minimum possible score. Where data did
not meet criteria for ceiling/floor effects, they were inspected for skew determined by visual inspection of data frequency
distributions. Test-retest reliability was determined from two-way mixed effects intraclass correlations between baseline
and 18-month follow-up test scores for each outcome measure. An intraclass correlation of >.50 was classified as indicating
acceptable test-retest reliability (Koo & Li, 2016). Remaining test outcome measures were then examined for redundancy by
computing the correlations between outcome measures of the same cognitive domain derived from the same neuropsychological
test. If two outcome measures from the same test showed a significant correlation, then one of the measures was subsequently
removed.

Table 1. Validity of cognitive assessments—outliers, ceiling and floor effects, missing data, skew, and test-retest reliability
Mean SD % Missing % Max score % Min score Skew Test-retest reliability (ICC)
Executive Function
Digit Span 18.19 3.95 0.5 0 0 Normal .81
FAS 42.14 11.05 1.8 — — Normal .88
Animals Names 40.43 7.75 1.8 — — Normal .79
Fruit Furniture Total 14.05 2.65 1.1 0 0 Normal .60
Fruit Furniture Switching 13.13 2.77 1.7 — — Normal .49
One Back 2.93 0.09 10.9 — — Normal .75
Stroop Interference 2.30 0.56 5.2 — — Positive .50
Episodic Memory
CVLT-II 1–5 52.90 9.93 0.8 0 0 Normal .8
CVLT-II Short Delay 11.50 2.95 0.8 8 0 Negative .79
CVLT-II Long Delay 12.19 2.78 1.5 10 0 Negative .79
CVLT-II False Positives 1.76 1.82 6.8 0 28 Positive .49
CVLT-II Recognition 14.97 1.25 1.7 45 0 Negative .49
Logical Memory 1 13.07 3.83 0.5 0 0 Normal .67
Logical Memory 2 11.67 3.80 1.4 0 0 Normal .70
RCFT Short Delay 16.89 5.63 0.3 0 0 Normal .79
RCFT Long Delay 16.58 5.65 0.5 0 0 Normal .78
RCFT Recognition 20.43 1.82 1.5 3 0 Negative .48
One Card Learning 1.07 0.13 11.1 — — Normal .74
Language
BNT Australian Version 28.58 1.38 7.3 30 0 Negative .70
BNT U.S. Version 28.3 1.70 16.3 28 0 Negative .73
Psychomotor Speed and Attention
Digit Symbol Coding 61.85 13.16 0.5 0 0 Normal .89
Stroop Dots 13.17 2.44 4.1 — — Normal .70
Stroop Words 17.19 3.22 4.9 — — Normal .81
Stroop Colors 30.16 7.27 4.9 — — Normal .78
Detection 2.51 0.10 12.3 — — Normal .59
Identification 2.71 0.05 11.1 — — Normal .73
Visuospatial Function
Clock 9.77 0.74 1.1 87 0 Negative .39
RCFT Copy 31.64 2.80 2.7 5 0 Negative .46
Note: CVLT-II = California Verbal Learning Test-Second Edition, RCFT = Rey Complex Figure Test, BNT = Boston Naming Test, ICC = Intraclass
Correlation Coefficient, FAS = Letter Fluency.
 K.D. Harrington et al. / Archives of Clinical Neuropsychology 32 (2017); 142–154 147

Results

Sample Characteristics. Demographic and clinical characteristics of the sample are summarized in Table 2. There were
slightly more women than men, average years of education was 12.5, IQ was estimated to be in the high average range, and
levels of depression and anxiety symptoms were uniformly low. Additionally, vital signs of the selected sample were within
subclinical ranges according to criteria from Bergman and colleagues (2007); systolic blood pressure M = 137.84 mmHg,
SD = 15.52, diastolic blood pressure M = 79.07 mmHg, SD = 10.08, and BMI M = 26.60 kg/m2, SD = 4.19. The medical
examination showed rates of systemic illness that were also low and in all cases were of subclinical severity and controlled

Downloaded from https://academic.oup.com/acn/article/32/2/142/2655685 by Universidad de Malaga user on 14 November 2023

medically; 22.1% hypertension, 6.5% diabetes, 2.0% skin cancer, 7.4% thyroid or parathyroid disease, 20.2% gastric com-
plaints (primarily reflux), 49.2% arthritis (primarily osteoarthritis), 2.1% kidney disease, 1.4% liver disease, and 2.6% neuro-
logical complaint (primarily migraines, nerve pain, or benign tremor).

Test Psychometrics. Table 1 summarizes the results of the psychometric investigations for each of the neuropsychological
outcome measures. Ceiling effects were identified for the Clock Drawing, CVLT-II Recognition, and BNT (Australian and
U.S. versions) outcome measures, and a floor effect was identified for the False Positive score on the CVLT-II Recognition
trial. Low test-retest reliability was identified for the Stroop Interference score, Fruit Furniture Switching score, and the Copy
and Recognition scores on the RCFT. Additionally, Stroop Dots (r = .47) and Colors (r = .55) trials, and CVLT-II Short
Delay trial (r = .78) were excluded as they were significantly correlated with other outcome measures from the same neuro-
psychological test.

Discussion

The exploratory analysis identified that many of the tests included in AIBL battery did not possess psychometric properties
optimal for investigation of the effects of age on cognition. A battery of neuropsychological tests whose outcome data were
not characterized by ceiling or floor effects, skew, or low reliability was identified (Table 3). This battery provided reliable
measures of learning, immediate and delayed recall aspects of episodic memory, psychomotor speed, attention, and executive
function that are appropriate and relevant for investigating age-related cognitive changes.

Study 2: The Effect of Age on Cognition

The aim of the second study was to characterize the effects of age on cognition in healthy older adults using the optimized
battery of tests identified in Study 1.

Neuropsychological Outcome Measures

Selected tests did not show floor or ceiling effects, substantial distributional skew, and had acceptable test-retest reliability
over 18 months. Table 3 summarizes the selected tests and corresponding cognitive domains.

Table 2. Median (range) demographic and clinical characteristics for each age group and total sample
Total 60–64 65–69 70–74 75–79 80–84 p
N 658 125 190 162 118 63
%Women 61.1 65.5 65.3 55.6 58.5 58.7 .28
% Post-secondary education 56.6 62.4 57.4 54.9 56.8 46.8 .36
Years of educationa 12.50 (2.95) 13.01 (3.01) 12.52 (2.96) 12.53 (2.82) 12.32 (3.08) 11.49 (2.75) .13
WTAR—U.S. normed 114 (46) 114 (31) 114 (34) 112 (32) 114 (28) 114 (39) .60
HADS-depression 2 (13) 1.5 (10) 2 (13) 2 (8) 2 (9) 3 (11) .08
HADS-anxiety 4 (15) 4 (13) 4 (13) 4 (14) 4 (14) 4 (15) .61
N present systemic illnesses 1 (5) 1 (4) 1 (4) 1 (4) 1 (4) 2 (5) .00
Note: WTAR = Wechsler Test of Adult Reading, HADS = Hospital Anxiety and Depression Scale.
a
Mean (SD) reported.
148 K.D. Harrington et al. / Archives of Clinical Neuropsychology 32 (2017); 142–154

Table 3. Analysis of variance results comparing age groups on neuropsychological outcome measures and group means and standard deviations
60–64 65–69 70–74 75–79 80–84 F p

N Mean SD N Mean SD N Mean SD N Mean SD N Mean SD

Speed and Attention
Digit Symbol Coding 125 68.78 12.95 188 64.80 11.37 162 60.36 12.19 117 57.02 13.23 63 52.08 10.89 28.20 .00
Stroop Wordsa 122 16.46 2.88 183 16.32 3.11 157 17.62 3.34 105 17.86 2.77 59 19.07 3.48 12.59 .00
Detectiona 107 2.51 0.09 166 2.51 0.09 143 2.51 0.10 104 2.53 0.09 57 2.50 0.10 1.57 .18
Identificationa 108 2.70 0.05 170 2.70 0.05 146 2.71 0.06 104 2.72 0.05 57 2.72 0.05 3.18 .01

Downloaded from https://academic.oup.com/acn/article/32/2/142/2655685 by Universidad de Malaga user on 14 November 2023

Executive Function
Digit Span 125 18.62 4.09 189 18.52 4.01 162 18.11 4.10 116 17.85 3.60 63 17.17 3.54 1.98 .10
Letter Fluency (FAS) 122 43.21 10.39 183 42.97 10.81 162 42.37 11.27 116 40.55 10.78 63 39.95 12.49 1.79 .13
Animals Names 124 42.10 8.03 185 41.39 7.25 162 40.28 7.98 113 38.61 7.42 62 37.89 7.47 5.54 .00
Fruit Furniture Total 125 14.71 2.72 187 14.43 2.63 161 14.17 2.51 116 13.09 2.52 62 13.06 2.53 9.42 .00
One Backa 109 2.91 0.08 170 2.93 0.09 145 2.94 0.08 108 2.95 0.08 54 2.97 0.09 5.27 .00
Learning/Immediate Recall
Logical Memory 1 125 13.65 3.72 189 13.63 3.71 162 12.85 3.59 117 12.21 4.24 62 12.37 3.88 3.95 .00
CVLT-II 1–5 124 55.98 9.04 189 55.42 9.76 161 51.95 9.33 116 49.95 10.10 63 47.11 8.99 15.59 .00
RCFT Short 125 18.38 5.42 189 17.54 5.86 162 17.13 5.30 117 15.47 5.40 63 14.03 5.12 9.27 .00
One Card Learninga 110 1.05 0.13 172 1.07 0.13 147 1.07 0.12 110 1.08 0.13 57 1.08 0.13 1.09 .36
Delayed Recall
Logical Memory 2 125 12.29 3.76 188 12.19 3.86 158 11.58 3.40 116 10.90 4.12 62 10.60 3.68 4.23 .00
CVLT-II Long Delay 125 12.79 2.64 188 12.72 2.75 159 11.97 2.66 114 11.69 2.86 62 10.87 2.71 8.12 .00
RCFT Long 125 18.13 5.31 188 17.24 6.07 162 16.86 5.28 117 15.07 5.29 63 13.56 5.01 10.19 .00
Note: CVLT-II = California Verbal Learning Test-Second Edition, RCFT = Rey Complex Figure Test.
a
Outcome measure is speed thus higher scores are indicative of worse performance.

Data Analysis

To compare cognitive performance at different ages, participants were grouped into 5-year age bands (60–64, 65–69,
70–74, 75–79, and 80–84). Scores on the WTAR were coded categorically according to descriptive IQ classification ranges
(Low Average = 80–89, Average = 90–110, High Average = 111–120, Well-Above Average/Superior = 121–130;
Wechsler, 1997).
To determine the effect of age on cognition, a series of one-way analysis of variance (ANOVA) were conducted to com-
pare mean performance across the five age groups on each of the selected neuropsychological outcome measures. Post hoc
Dunnett tests were then conducted to compare the performance of the 60- to 64-year-old group with each of the other age
groups. Finally, to determine the magnitude of the differences between age group performance on each of the outcome mea-
sures Hedges’ g effect size measure of standardized mean difference was calculated to express the differences in SD units.
To explore the extent to which gender and IQ affected the relationship between age and cognitive performance, a series of
multiple regressions were conducted. Age (entered as a continuous variable), gender, and WTAR categorical classification
were entered as independent variables, and performance on each of the selected neuropsychological outcome measures
entered as the dependent variable.

Results

Sample characteristics. The demographic and clinical characteristics of each age group are summarized in Table 2. No signif-
icant differences were identified between the age groups for the proportion of women, level of education, estimated IQ, or le-
vels of depression and anxiety symptoms. However, the 80- to 84-year-old group reported a slightly greater number of
comorbid systemic illnesses than the other groups for their present health status on the medical history interview. Importantly,
all systemic illnesses reported were of subclinical severity and controlled medically such that they were unlikely to affect cog-
nitive function, hence analyses proceeded without any adjustment for covariates.

Effect of age on cognition. There was a significant effect of age on performance on each neuropsychological test except
the Digit Span, FAS, One Card Learning, and Detection tests (see Table 3 for summary). Fig. 2 summarizes the results of
the post hoc Dunnett tests and the magnitude of the differences in cognitive performance between the 60- to 64-year-old
group and each of the other age groups. Age differences were not observed between the 60- to 64-year-old group and the
 K.D. Harrington et al. / Archives of Clinical Neuropsychology 32 (2017); 142–154 149

Downloaded from https://academic.oup.com/acn/article/32/2/142/2655685 by Universidad de Malaga user on 14 November 2023

Fig. 2. Magnitude of mean difference (Hedges’ g) in cognitive performance between the 60- to 64-year-old group and each of the other age groups. *p < .05.
CVLT-II = California Verbal Learning Test-Second Edition, RCFT = Rey Complex Figure Test.

65- to 69-year-old group on any measure except Digit Symbol Coding, whereas the 75–79 and 80–84 year groups differed
from the 60- to 64-year-old group on every measure except Identification and Logical Memory 1.

Effect of gender and IQ on age-related cognitive change. Multiple regression outcomes are summarized in Table 4.
Significant age effects were observed for almost all cognitive measures, with age accounting for the greatest variance in per-
formance on Digit Symbol Coding, Stroop Words, CVLT-II Trials 1–5, RCFT Short and Long Delay, Fruit Furniture Total
Score, One Back, and One Card Learning. Gender and IQ were also significant predictors of performance on most measures
despite there being no differences between age groups on these demographic variables.

Discussion

The first hypothesis that older age would be associated with worse cognitive performance on all neuropsychological tests
was supported partially. Moderate to large (Hedges’ g = −0.34 to −1.34) age-related decreases in performance were
observed for each neuropsychological test except the Digit Span, FAS, One Card Learning, and Detection tests (Fig. 2).
For tests where no age effect was identified the magnitude of differences between the youngest and oldest cohorts was
small (Hedges’ g = −0.01 to −0.36). As the Hedges’ g effect size is equivalent to expressing impairment in terms of SD units, it
can be compared directly with estimates of effects observed in prior studies of aging. Prior cross-sectional comparisons show
age-related cognitive impairment that was systematically greater than that observed here (e.g., 0.5–1.5 SD unit decrease from age
60 to 80) for processing speed, memory, reasoning, and spatial ability (Schaie, 2005; van Hooren et al., 2007; Salthouse, 2010;
Bonsang & Dohmen, 2015). Furthermore, unlike in these previous studies of cognitive aging, the effect of age on cognitive per-
formance was not consistent across all neuropsychological measures in the current study. One hypothesis is that the smaller and
more specific effect of age on cognition observed in the current study was because this study took great care to exclude
150 K.D. Harrington et al. / Archives of Clinical Neuropsychology 32 (2017); 142–154

Table 4. Multiple regression results for neuropsychological outcome measures with age, gender and IQ as predictors
R2 Partial r

Age Gender WTAR

Speed and Attention
Digit Symbol Coding .19** −.38** −.06 .22**
Stroop Wordsa .11** .26** .02 −.23**
Detection .01 .07 −.02 −.05
Identification .04** .14** −.03 −.15**

Downloaded from https://academic.oup.com/acn/article/32/2/142/2655685 by Universidad de Malaga user on 14 November 2023

Executive Function
Digit Span .13** −.13** .15** .32**
Letter Fluency (FAS) .15** −.09* −.01 .38**
Animals Names .09** −.18** −.07 .22**
Fruit Furniture Total .12** −.24** −.23** .14**
One Backa .06** .19** −.08 −.13**
Learning/Immediate Recall
Logical Memory 1 .07** −.15** −.09* .21**
CVLT-II 1–5 .19** −.30** −.27** .24**
RCFT Short Delay .12** −.25** .22** .16**
One Card Learning .04** −.15** −.03 .14**
Delayed Recall
Logical Memory 2 .08** −.16** −.09* .22**
CVLT-II Long Delay .13** −.21** −.25** .19**
RCFT Long Delay .12** −.26** .21** .15**
Note: *p < .05, **p < .01. WTAR = Wechsler Test of Adult Reading, CVLT-II = California Verbal Learning Test-Second Edition, RCFT = Rey Complex
Figure Test.
a
Outcome measure is speed thus higher scores are indicative of worse performance.

individuals with prodromal CNS disease and uncontrolled systemic illness, whereas most other studies have not. The small effect
of age on cognition was confirmed in multivariate analyses assessing the extent to which age explained variance beyond gender
and IQ. With these additional factors added to the models, the variance explained by age alone was very small (e.g., partial
r = 0 to −.38; Table 4). Similarly, education level and gender have been observed to have a substantial effect on cognitive per-
formance independent of age effects, in a large community-based sample of older adults (aged 64–81 years) who did not have
significant medical conditions and scored above 24 on the MMSE at enrollment (van Hooren et al., 2007).
These data indicate that while complex psychomotor function, categorical fluency, verbal learning, and verbal and visual
memory showed age-related impairment, in each case the magnitude of this was small when the contributions of gender and
IQ were controlled statistically. Additionally, little or no age-related impairment was observed for working memory, phone-
mic fluency, learning visual information, and reaction time (Table 4). Importantly, the effect of age on neuropsychological
test performance was examined using tests with optimal psychometric properties and in a robust sample, without uncontrolled
systemic medical illness and with a very low probability of prodromal CNS disease. Under these conditions, the effect of age
on most aspects of cognition is small. The hypothesis that previous studies of cognitive aging have overestimated age-related
cognitive impairment due to lack of precision in their definitions of healthy aging is theoretically consistent with the use of
robust normative data in neuropsychological contexts. In these contexts, it is acknowledged that inclusion of individuals with
suboptimal health, or indeed preclinical neurodegenerative disease, in normative samples will decrease the sensitivity of normal
data ranges, derived from those samples, to identify true cognitive impairment in older people at risk of CNS disease. For exam-
ple, inclusion of data from healthy older adults who later progress to dementia in normative estimates causes age-based estimates
of cognitive performance to be lower and to have greater variability (Sliwinski et al., 1996; De Santi et al., 2008; Holtzer et al.,
2008; Pedraza et al., 2010; Grober, Mowrey, Katz, Derby, & Lipton, 2015; Clark et al., 2016). Thus, the development and appli-
cation of robust normative data for neuropsychology is consistent with the hypothesis that in cross-sectional studies of aging the
effects of age on cognition have been overestimated due to the inclusion of individuals who are not healthy.
The second hypothesis that performance on tests of psychomotor speed and episodic memory would show the greatest age-
related impairment was supported partially. Although the largest age-related impairment in cognition was observed for tests of psy-
chomotor speed and episodic memory, these effects were not consistent across all neuropsychological tests within these domains
(see Fig. 2). For example, when psychomotor speed was measured using the Digit Symbol Coding test the magnitude of difference
between the youngest and oldest groups was substantial (Hedges’ g = −1.34) in line with estimates derived from multiple studies
using similar tests (van Hooren et al., 2007; Salthouse, 2010). In comparison there were no, or only small (Hedges’ g < −0.44),
effects of age observed when psychomotor speed was measured with tests of simple or choice reaction time (i.e., Detection and
 K.D. Harrington et al. / Archives of Clinical Neuropsychology 32 (2017); 142–154 151

Identification tasks). This is consistent with prior observations in a large population-based sample of older adults that there is stabil-
ity of simple and choice reaction time over 13 years (Deary, Allerhand, & Der, 2009). Similarly, for episodic memory the magni-
tude of difference between the 60- and 80-year-old groups was greatest for verbal word-list learning (Hedges’ g = −0.97) and
delayed recall (i.e., CVLT; Hedges’ g = −0.71); while no, or only small, effects of age were observed for immediate and delayed
recall of a short story (i.e., Logical Memory 1 and 2, Hedges’ g < −0.45). Word-list learning and recall has been observed to be
more sensitive to early cognitive changes associated with AD and PD than story recall (Rabin et al., 2009; Zahodne et al., 2011).
Thus, word-list learning may be more sensitive than story recall to changes in memory generally.
One way to integrate the different effects of age on performance on different neuropsychological tests measuring the same

Downloaded from https://academic.oup.com/acn/article/32/2/142/2655685 by Universidad de Malaga user on 14 November 2023

cognitive domain is to consider these tests according to their complexity. For example, for the domain of psychomotor speed,
age-related impairment was much greater for tests that required the integration of multiple cognitive operations (i.e., Digit
Symbol Coding) than for measures that required only simple reflexive responses (i.e., Detection and Identification). Similarly,
for episodic memory, age-related impairment in cognition was greater for tests with less contextual detail or support from gist
(i.e., CVLT-II) than those where gist or semantic scaffolding supported encoding (i.e., Logical Memory). Thus, one hypothe-
sis arising from this study is that age-related cognitive impairment is dependent not just on the domain of cognition, but also
on the complexity of information processing required within that domain. Data from prior studies of CN adults support this
hypothesis, reporting larger age effects for tasks that require more active and effortful processing than for simple tasks of the
same cognitive process (van Hooren et al., 2007; Harris, Eckert, Ahlstrom, & Dubno, 2010; Frey, Mata, & Hertwig, 2015).

General Discussion

When considered together, the results from this study suggest that age-related cognitive impairment may be of a smaller magni-
tude than reported previously (e.g., the relationship between age and speed, reasoning, and memory performance observed to be
r = −.47, −.48, and −.43, respectively [Salthouse, 2004], or about 1–1.5 SD difference in performance between 60 and 80 years
old [Schaie, 2005; van Hooren et al., 2007; Salthouse, 2010; Bonsang & Dohmen, 2015]). Prior cross-sectional studies may have
overestimated the magnitude of cognitive impairment associated with aging due to their use of inclusion/exclusion criteria that
lack precision, leading to the assessment of individuals with severe or uncontrolled systemic illness or even prodromal CNS dis-
ease, and the subsequent confounding of the effect of these health factors with aging. Notably, despite rigorous screening for cog-
nitive and physical health prior to entry, approximately 10% of the AIBL sample, classified as CN initially, went on to be
diagnosed with MCI or dementia over the next 72 months. A further 6% were identified to have had a vascular event, such as a
stroke, or to develop a significant medical condition such as cancer (see Fig. 1). These rates of subclinical CNS and systemic dis-
ease are likely to be greater in other healthy aging samples that applied less rigorous criteria at screening than the AIBL study
(e.g., Carlesimo et al., 1998; Bates & Wolbers, 2014; Bonsang & Dohmen, 2015; Brown, Johnson, Sohl, & Dumas, 2015).
The results of this study also offer estimates of the normal effects of age on neuropsychological tests used commonly in
decision-making about neurodegenerative disease, derived from a group of well-characterized older adults without uncontrolled
systemic medical illness or prodromal CNS disease. Specifically, Table 3 provides estimates of group mean performance and vari-
ation for each age quintile between 60 and 84 years. These estimates of the effect of aging on cognition, developed from a cogni-
tively healthy group and based on neuropsychological tests that have optimal characteristics for measurement of cognitive change,
may be of use to neuropsychologists engaged in clinical assessment and diagnosis. The methods used to refine these estimates by
excluding data from individuals who had severe or uncontrolled systemic illness or prodromal CNS disease are equivalent to those
used in the generation of robust neuropsychological data. Thus, the age-based estimates of normal cognitive performance pre-
sented here could facilitate greater sensitivity to cognitive impairment in older adults at risk of CNS injury, disease, or disorder
than normative data used currently (De Santi et al., 2008; Grober et al., 2015; Clark et al., 2016) .
There are some caveats for the interpretation of the results of the present study. First, this is an experimental convenience
sample and as such may not be representative of the broader aging population. The rigorous inclusion criteria applied here
ensured the physical and cognitive health of the sample and rendered it ideal for developing models of cognitive aging, inde-
pendent of the effects of other health or social factors. However, one issue arising from the methodology applied here to iden-
tify healthy aging, as well as acceptance of robust normative samples in neuropsychology more generally, is whether such
samples should be considered to reflect normal aging, or whether they should be defined as super aging or successful aging.
Although this idea is compelling, there remains an absence of any clear neuropsychological definition or criteria for super or
successful aging. Therefore, in this context we suggest to focus on the concept of robust estimates of cognitive aging, until
such criteria for super or successful aging are developed. Thus, the data averages and ranges developed from the current sam-
ple (Table 3) may be of clinical utility to neuropsychologists. Second, it is known that systemic illness (e.g., hypertension,
dyslipidaemia, or diabetes) occurring during early- or mid-life may affect CNS function and cognitive ability in later life
152 K.D. Harrington et al. / Archives of Clinical Neuropsychology 32 (2017); 142–154

(van Vliet, 2012; Gottesman et al., 2014; Rawlings et al., 2014). Although these effects have not been accounted for in the
present study, it is unlikely that they will have substantially influenced estimates of age-related cognitive impairment in the
present sample as all individuals provided a detailed medical history and free universal healthcare and subsidized medications
have been available in Australia for the past 60 years. Third, the selection of neuropsychological tests in the present study
was constrained by the aims of the AIBL study rather than by theoretical models of aging. However, the tests selected were
chosen for their sensitivity to dementia, and therefore they will be familiar and commonly used by neuropsychologists.
Finally, the analyses used to consider the effect of age on cognitive function in the present study were cross-sectional design.
Accordingly, no inferences regarding change in cognitive function over time can be made. Additionally, it is known that esti-

Downloaded from https://academic.oup.com/acn/article/32/2/142/2655685 by Universidad de Malaga user on 14 November 2023

mates of the effect of age on cognition based on studies with cross-sectional designs can be influenced by cohort effects
(Schaie, 2005; Salthouse, 2014). These reflect variability in samples due to age-related differences in environmental factors
such as access to healthcare and education. In the current study, we have sought to mitigate the potential for cohort effects to
bias results by selecting IQ rather than education to characterize the premorbid ability of the sample, and conducting the
research in a context where there has been universal free healthcare access for the past 60 years. Although these aspects of the
sample may not entirely overcome cohort effects, they do minimize the major effects. However, further investigation of cogni-
tive aging in the absence of prodromal CNS disease and severe systemic illness with prospective design is now necessary.

Conclusion

These caveats notwithstanding, the results from the present study provide an important indication of the nature and magni-
tude of age-related cognitive impairment in healthy older individuals. Notably, the impact of aging does not appear to be
ubiquitous across cognitive domains, but in fact may be limited to higher order complex functions and of a smaller magnitude
occurring later than previously reported. In terms of clinical decision-making, the estimates of age-related cognitive
impairment from the present study provide objective reference ranges of performance on specific neuropsychological tests for
very healthy older adults (see Table 3 for normative means and SD derived for these). Thus, cognitive impairment falling
within these ranges likely reflects true cognitive aging.

Funding

Ms Harrington received a PhD Scholarship co-funded by the Alzheimer’s Australia Dementia Research Foundation and the
Florey Institute of Neuroscience and Mental Health. The authors acknowledge the financial support of the CRC for Mental
Health. The Cooperative Research Centre (CRC) program is an Australian Government Initiative. Funding for the AIBL study
was provided in part by the study partners (Australian Commonwealth Scientific Industrial and research Organization [CSIRO],
Edith Cowan University [ECU], Mental Health Research Institute [MHRI], Alzheimer’s Australia [AA], National Ageing
Research Institute [NARI], Austin Health, CogState Ltd., Hollywood Private Hospital, Sir Charles Gardner Hospital). The study
also received support from the National Health and Medical Research Council (NHMRC) and the Dementia Collaborative
Research Centres program (DCRC2), as well as ongoing funding from the Science and Industry Endowment Fund (SIEF).

Acknowledgements

Alzheimer’s Australia (Victoria and Western Australia) assisted with promotion of the AIBL study and the screening of
telephone calls from volunteers. The AIBL team wishes to thank the clinicians who referred patients with AD to the study:
Associate Professor Brian Chambers, Professor Edmond Chiu, Dr Roger Clarnette, Associate Professor David Darby,
Dr Mary Davison, Dr John Drago, Dr Peter Drysdale, Dr Jacqueline Gilbert, Dr Kwang Lim, Professor Nicola
Lautenschlager, Dr Dina LoGiudice, Dr Peter McCardle, Dr Steve McFarlane, Dr Alastair Mander, Dr John Merory,
Professor Daniel O’Connor, Dr Ron Scholes, Dr Mathew Samuel, Dr Darshan Trivedi, and Associate Professor Michael
Woodward. We thank all those who participated in the study for their commitment and dedication to helping advance research
into the early detection and causation of AD.

Conflict of Interest

KDH, JH, SR-S, JR, and OS report no disclosures. PM and YYL are employees of Cogstate Ltd. DA has served on scien-
tific advisory boards for Novartis, Eli Lilly, Janssen, and Pfizer Inc. CLM is an advisor to Prana Biotechnology Ltd and a con-
sultant to Eli Lilly.
 K.D. Harrington et al. / Archives of Clinical Neuropsychology 32 (2017); 142–154 153

References

Agrell, B., & Dehlin, O. (1998). The clock-drawing test. Age and Ageing, 27, 399–403. doi:10.1093/ageing/27.3.399.
Bates, S. L., & Wolbers, T (2014). How cognitive aging affects multisensory integration of navigational cues. Neurobiology of Aging, 35 (12), 2761–2769.
doi:10.1016/j.neurobiolaging.2014.04.003.
Bergman, I., Blomberg, M., & Almkvist, O. (2007). The importance of impaired physical health and age in normal cognitive aging. Scandinavian Journal of
Psychology, 48 (2), 115–125.
Bonsang, E., & Dohmen, T. (2015). Risk attitude and cognitive aging. Journal of Economic Behavior & Organization, 112, 112–126. doi:10.1016/j.jebo.2015.01.004.
Brown, C. L., Gibbons, L. E., Kennison, R. F., Robitaille, A., Lindwall, M., Mitchell, M. B., et al. (2012). Social activity and cognitive functioning over
time: A coordinated analysis of four longitudinal studies. Journal of Aging Research, 1–12. doi:10.1155/2012/287438.

Downloaded from https://academic.oup.com/acn/article/32/2/142/2655685 by Universidad de Malaga user on 14 November 2023

Brown, S. W., Johnson, T. M., Sohl, M. E., & Dumas, M. K. (2015). Executive attentional resources in timing: Effects of inhibitory control and cognitive
aging. Journal of Experimental Psychology, Human Perception & Performance, 41, 1063–1083. doi:10.1037/xhp0000078.
Carlesimo, G. A., Mauri, M., Graceffa, A. M., Fadda, L., Loasses, A., Lorusso, S., et al. (1998). Memory performances in young, elderly, and very old
healthy individuals versus patients with Alzheimer’s disease: Evidence for discontinuity between normal and pathological aging. Journal of Clinical and
Experimental Neuropsychology, 20, 14–29.
Clark, L. R., Koscik, R. L., Nicholas, C. R., Okonkwo, O. C., Engelman, C. D., Bratzke, L. C., et al. (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. doi:10.1093/arclin/acw024.
Clouston, S. A. P., Glymour, M. M., & Terrera, G. M. (2015). Cognitive & Behavioral Assessment: Educational inequalities in aging-related declines in fluid
cognition and the onset of cognitive pathology. Alzheimer’s & Dementia: Diagnosis, Assessment & Disease Monitoring, 1, 303–310. 10.1016/j.dadm.
2015.06.001.
De Santi, S., Pirraglia, E., Barr, W., Babb, J., Williams, S., Rogers, K., et al. (2008). Robust and conventional neuropsychological norms: diagnosis and pre-
diction of age-related cognitive decline. Neuropsychology, 22, 469–484. doi:10.1037/0894-4105.22.4.469.
Deary, I. J., Allerhand, M., & Der, G. (2009). Smarter in middle age, faster in old age: a cross-lagged panel analysis of reaction time and cognitive ability
over 13 years in the West of Scotland Twenty-07 Study. Psychology And Aging, 24, 40–47. doi:10.1037/a0014442.
Deary, I. J., Corley, J., Gow, A. J., Harris, S. E., Houlihan, L. M., Marioni, R. E., et al. (2009). Age-associated cognitive decline. British Medical Bulletin,
92, 135–152. doi:10.1093/bmb/ldp033.
Deary, I. J., Johnson, W., & Starr, J. M. (2010). Are processing speed tasks biomarkers of cognitive aging? Psychology And Aging, 25, 219–228. doi:10.
1037/a0017750.
Delis, D., Kaplan, E., & Kramer, J. H. (2001). The Delis-Kaplan Executive Function System (D-KEFS). San Antonio TX: Psychological Corporation.
Delis, D., Kramer, J., Kaplan, E., & Ober, B. (2000). California Verbal Learning Test (2nd ed.). San Antonio, TX: Psychological Corporation.
Ellis, K. A., Bush, A. I., Darby, D., De Fazio, D., Foster, J., Hudson, P., et al. (2009). The Australian Imaging, Biomarkers and Lifestyle (AIBL) study of
aging: methodology and baseline characteristics of 1112 individuals recruited for a longitudinal study of Alzheimer’s disease. International
Psychogeriatrics, 21, 672–687. doi:doi:10.1017/S1041610209009405.
Frey, R., Mata, R., & Hertwig, R. (2015). The role of cognitive abilities in decisions from experience: Age differences emerge as a function of choice set
size. Cognition, 142, 60–80. doi:http://dx.doi.org/10.1016/j.cognition.2015.05.004.
Gerstorf, D., Ram, N., Hoppmann, C., Willis, S. L., & Schaie, K. W. (2011). Cohort Differences in Cognitive Aging and Terminal Decline in the Seattle
Longitudinal Study. Developmental Psychology, 47, 1026–1041.
Gottesman, R. F., Schneider, A. L. C., Albert, M., Alonso, A., Bandeen-Roche, K., Coker, L., et al. (2014). Midlife hypertension and 20-year cognitive
change: The Atherosclerosis Risk in Communities Neurocognitive Study. JAMA Neurology, 71, 1218–1227. doi:10.1001/jamaneurol.2014.1646.
Grober, E., Mowrey, W., Katz, M., Derby, C., & Lipton, R. B. (2015). Conventional and robust norming in identifying preclinical dementia. Journal of
Clinical and Experimental Neuropsychology, 37 (10), 1098. doi:10.1080/13803395.2015.1078779.
Harris, K. C., Eckert, M. A., Ahlstrom, J. B., & Dubno, J. R. (2010). Age-related differences in gap detection: Effects of task difficulty and cognitive ability.
Hearing research, 264, 21–29. doi:10.1016/j.heares.2009.09.017.
Heine, C., & Browning, C. (2015). Dual Sensory Loss in Older Adults: A Systematic. Gerontologist, 55, 913.
Holtzer, R., Goldin, Y., Zimmerman, M., Katz, M., Buschke, H., & Lipton, R. B. (2008). Robust norms for selected neuropsychological tests in older adults.
Archives of Clinical Neuropsychology, 23, 531–541. doi:http://dx.doi.org/10.1016/j.acn.2008.05.004.
Jefferson, A. L., Wong, S., Gracer, T. S., Ozonoff, A., Green, R. C., & Stern, R. A. (2007). Geriatric Performance on an Abbreviated Version of the Boston
Naming Test. Applied neuropsychology, 14, 215–223. doi:10.1080/09084280701509166.
Koo, T. K., & Li, M. Y. (2016). Cracking the Code: Providing Insight Into the Fundamentals of Research and Evidence-Based Practice: A Guideline of
Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research. Journal of Chiropractic Medicine, 15, 155–163. doi:10.1016/j.jcm.
2016.02.012.
Lezak, M. D. (2012). Neuropsychological assessment (5th ed.). Oxford; New York: Oxford University Press. Muriel D. Lezak. [et al.] c2012.
Lim, Y. Y., Ellis, K. A., Harrington, K., Ames, D., Martins, R. N., Masters, C. L., et al. (2012). Use of the CogState Brief Battery in the assessment of
Alzheimer’s disease related cognitive impairment in the Australian Imaging, Biomarkers and Lifestyle (AIBL) study. Journal of clinical and experimental
neuropsychology, 34, 345–358.
Lindenberger, U. (2014). Human cognitive aging: corriger la fortune? Science (New York, N.Y.), 346 (6209), 572–578. doi:10.1126/science.1254403.
McKhann, G., Drachman, D., Folstein, M., Katzman, R., Price, D., & Stadlan, E. M. (1984). Clinical diagnosis of Alzheimer’s disease: report of the
NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology, 34,
939–944.
Melis, R., Marengoni, A., Angleman, S., & Fratiglioni, L. (2014). Incidence and Predictors of Multimorbidity in the Elderly: A Population-Based
Longitudinal Study. PloS one, 9, e103120. doi:10.1371/journal.pone.0103120.
Meyers, J. E., & Meyers, K. R. (1995). Rey Complex Figure Test and Recognition Trial: Professional Manual: Psychological Assessment Resource, Inc.
154 K.D. Harrington et al. / Archives of Clinical Neuropsychology 32 (2017); 142–154

Pedraza, O., Lucas, J. A., Smith, G. E., Petersen, R. C., Graff-Radford, N. R., & Ivnik, R. J. (2010). Robust and Expanded Norms for the Dementia Rating
Scale. Archives of Clinical Neuropsychology, 25, 347–358. doi:10.1093/arclin/acq030.
Petersen, R. C., Aisen, P. S., Beckett, L. A., Donohue, M. C., Gamst, A. C., Harvey, D. J., et al. (2010). Alzheimer’s Disease Neuroimaging Initiative
(ADNI): Clinical characterization. Neurology, 74, 201–209. doi:10.1212/WNL.0b013e3181cb3e25.
Rabin, L. A., Paré, N., Saykin, A. J., Brown, M. J., Wishart, H. A., Flashman, L. A., et al. (2009). Differential Memory Test Sensitivity for Diagnosing
Amnestic Mild Cognitive Impairment and Predicting Conversion to Alzheimer’s Disease. Neuropsychology, development, and cognition. Section B,
Aging, neuropsychology and cognition, 16, 357–376. doi:10.1080/13825580902825220.
Rantakokko, M., Mänty, M., & Rantanen, T. (2013). Mobility decline in old age. Exercise And Sport Sciences Reviews, 41, 19–25. doi:10.1097/JES.
0b013e3182556f1e.
Rawlings, A. M., Sharrett, A. R., Schneider, A. L. C., Coresh, J., Albert, M., Couper, D., et al. (2014). Diabetes in midlife and cognitive change over 20

Downloaded from https://academic.oup.com/acn/article/32/2/142/2655685 by Universidad de Malaga user on 14 November 2023

years: the Atherosclerosis Risk in Communities Neurocognitive Study. Annals of internal medicine, 161, 785–793. doi:10.7326/M14-0737.
Roberts, R. O., Geda, Y. E., Knopman, D. S., Cha, R. H., Pankratz, V. S., Boeve, B. F., et al. (2008). The Mayo Clinic Study of Aging: Design and
Sampling, Participation, Baseline Measures and Sample Characteristics. Neuroepidemiology, 30, 58–69. doi:10.1159/000115751.
Salthouse, T. A. (2004). What and When of Cognitive Aging. Current Directions in Psychological Science, 13 (4), 140–144. doi:10.1111/j.0963-7214.2004.
00293.x.
Salthouse, T. A. (2010). Selective review of cognitive aging. Journal of the International Neuropsychological Society: JINS, 16, 754–760. doi:10.1017/
S1355617710000706.
Salthouse, T. A. (2014). Why are there different age relations in cross-sectional and longitudinal comparisons of cognitive functioning? Curr Dir Psychol Sci,
23, 252–256. doi:10.1177/0963721414535212.
Saxton, J., Raclif, G., Munro, C. A., Coffey, E. C., Becker, J. T., Fried, L., et al. (2000). Normative data on the Boston Naming Test and two equivalent
30-item short forms. Clinical Neuropsychology, 14, 526–534.
Schaie, K. W. (2005). What can we learn from longitudinal studies of adult development? Research in Human Development, 2, 133–158. doi:10.1207/
s15427617rhd0203_4.
Sliwinski, M., Lipton, R. B., Buschke, H., & Stewart, W. (1996). The effects of preclinical dementia on estimates of normal cognitive functioning in aging.
The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 51B, P217–P225. doi:10.1093/geronb/51B.4.P217.
Storandt, M., & Morris, J. C. (2010). Ascertainment bias in the clinical diagnosis of Alzheimer’s disease. Archives of Neurology, 67, 1364–1369. doi:10.
1001/archneurol.2010.272.
Strauss, E., Sherman, & Spreen, O. (2006). A compendium of neuropsychological tests: Administration, norms, and commentary (3rd ed.). New York: Oxford
University Press.
van Hooren, S. A. H., Valentijn, A. M., Bosma, H., Ponds, R. W. H. M., van Boxtel, M. P. J., & Jolles, J. (2007). Cognitive functioning in healthy older
adults aged 64–81: A cohort study into the effects of age, sex, and education. Aging, Neuropsychology, and Cognition, 14, 40–54. doi:10.1080/
138255890969483.
van Vliet, P. (2012). Cholesterol and late-life cognitive decline. Journal Of Alzheimer’s Disease: JAD, 30, S147–S162. doi:10.3233/JAD-2011–111028.
Wechsler, D. (1987). Wechsler Memory Scale-Revised. New York: Psychological Corporation.
Wechsler, D. (1997). Wechsler Adult Intelligence Scale (WAIS-III) (3rd ed.). San Antonio, TX: Psychological Corporation.
Wechsler, D. (2001). Wechsler test of adult reading: Examiner’s manual. San Antonio, TX: Psychological Corporation.
Weintraub, D., Simuni, T., Caspell-Garcia, C., Coffey, C., Lasch, S., Siderowf, A., et al. (2015). Cognitive performance and neuropsychiatric symptoms in
early, untreated Parkinson’s disease. Movement Disorders, 30 (7), 919. doi:10.1002/mds.26170.
Winblad, B., Palmer, K., Kivipelto, M., Jelic, V., Fratiglioni, L., Wahlund, L. O., et al. (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.
Zahodne, L. B., Bowers, D., Price, C. C., Bauer, R. M., Nisenzon, A., Foote, K. D., et al. (2011). The case for testing memory with both stories and word lists
prior to DBS surgery for Parkinson’s disease. The Clinical Neuropsychologist, 25, 348–358. doi:10.1080/13854046.2011.562869.
Zigmond, A. S., & Snaith, R. P. (1983). The Hospital Anxiety and Depression Scale. Acta Psychiatrica Scandinavica, 67, 361–370. doi:10.1111/j.1600-0447.
1983.tb09716.x.