Developmental Psychology
2013, Vol. 50, No. 1, 000

© 2013 American Psychological Association
0012-1649/13/$12.00 DOI: 10.1037/a0034913

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

The Malleability of Working Memory and Visuospatial Skills:
A Randomized Controlled Study in Older Adults
Hana Stepankova

Jiri Lukavsky

Prague Psychiatric Center, Prague, Czech Republic, and Charles
University in Prague

Prague Psychiatric Center, Prague, Czech Republic, and
Academy of Sciences of the Czech Republic, Prague

Martin Buschkuehl

Miloslav Kopecek and Daniela Ripova

University of California, Irvine, and MIND Research Institute,
Irvine, California

Prague Psychiatric Center, Prague, Czech Republic

Susanne M. Jaeggi
University of California, Irvine
There is accumulating evidence that training on working memory (WM) generalizes to other nontrained
domains, and there are reports of transfer effects extending as far as to measures of fluid intelligence.
Although there have been several demonstrations of such transfer effects in young adults and children,
they have been difficult to demonstrate in older adults. In this study, we investigated the generalizing
effects of an adaptive WM intervention on nontrained measures of WM and visuospatial skills. We
randomly assigned healthy older adults to train on a verbal n-back task over the course of a month for
either 10 or 20 sessions. Their performance change was compared with that of a control group. Our
results revealed reliable group effects in nontrained standard clinical measures of WM and visuospatial
skills in that both training groups outperformed the control group. We also observed a dose–response
effect, that is, a positive relationship between training frequency and the gain in visuospatial skills; this
finding was further confirmed by a positive correlation between training improvement and transfer. The
improvements in visuospatial skills emerged even though the intervention was restricted to the verbal
domain. Our work has important implications in that our data provide further evidence for plasticity of
cognitive functions in old age.
Keywords: aging, plasticity, transfer, dose–response effect, n-back

tions and processing speed, but it also extends to higher cognitive
functions such as executive control and fluid intelligence (Gf;
Baltes & Lindenberger, 1997; Bugg, Zook, DeLosh, Davalos, &
Davis, 2006; Park et al., 2002; Ryan, Sattler, & Lopez, 2000;
Salthouse, 1996), and, ultimately, everyday life functioning (Alzheimer’s Association, 2011). Therefore, it would be highly beneficial if there were means to prevent cognitive decline or even
improve old adults’ cognitive functions. In the current study, we
investigated whether a focused cognitive intervention targeting
working memory (WM) has the potential to serve as such a means
in a healthy young– old adult sample.
There have been numerous attempts to improve higher cognitive
functions in older adults, and in particular, researchers have been
focusing on improving fluid reasoning or memory by having
participants learn and implement cognitive strategies (e.g. Ball et
al., 2002; Schaie & Willis, 1986; Verhaeghen, Marcoen, & Goossens, 1992). However, although it has been shown that older adults
are able to considerably increase performance with practice, the
success of those studies has often been limited. That is, in many
cases, the generalizing effects were restricted to tasks that were
closely related to the trained domain (e.g. Hayslip, 1989; Zelinski,
2009). Nevertheless, in recent years and especially with the advent
of computerized and sophisticated adaptive intervention designs,

For ages, older adults have been respected for their acquired
expertise, knowledge, and wisdom (Rowley & Slack, 2008). Nevertheless, there is a decline in certain cognitive functions during
ontogenetic aging which is most apparent in basic sensory func-

Hana Stepankova, Prague Psychiatric Center and the Department of
Psychology, Faculty of Arts, Charles University in Prague, Prague, Czech
Republic; Jiri Lukavsky, Prague Psychiatric Center and the Institute of
Psychology, Academy of Sciences of the Czech Republic, Prague, Czech
Republic; Martin Buschkuehl, School of Education, University of California, Irvine, and the MIND Research Institute, Irvine, California; Miloslav
Kopecek and Daniela Ripova, Prague Psychiatric Center; Susanne M.
Jaeggi, School of Education, University of California, Irvine.
This work was supported by Grant MH CZ-DRO: PCP 00023752 from
the Prague Psychiatric Center and by the Programme for the Development
of Fields of Study at Charles University, No. P03 Development of Psychological Sciences, Sub-Programme Clinical Psychology. Susanne M.
Jaeggi and Martin Buschkuehl received support from the Office of Naval
Research as well as from the U.S. Institute for Educational Sciences.
Correspondence concerning this article should be addressed to Susanne
M. Jaeggi, School of Education, 3452 Education, University of California,
Irvine, Irvine, CA 92697-5500. E-mail: smjaeggi@uci.edu
1

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

2

STEPANKOVA ET AL.

the focus of training research has shifted towards “process-based”
interventions (Lustig, Shah, Seidler, & Reuter-Lorenz, 2009). For
example, there has been accumulating evidence that training on
WM can improve skills related to Gf (see, e.g., Buschkuehl &
Jaeggi, 2010; Morrison & Chein, 2011; Jaeggi, Buschkuehl, Shah,
& Jonides, 2013, for recent reviews).
WM is commonly described as a set of mechanisms capable of
retaining a limited amount of information in an active state for the
use in ongoing cognitive tasks (e.g. Cowan et al., 2005; Jonides et
al., 2008). Gf in turn is described as our general ability to understand and reason inductively about novel and abstract concepts,
that is, our ability to solve new or unfamiliar problems independently of previous specific practice or experience (Cattell, 1971;
Gottfredson, 1997). It is assumed that WM is one of the mechanisms that underlies individual differences in higher cognitive
functions, such as Gf (e.g., Oberauer, Schulze, Wilhelm, & Süss,
2005). Furthermore, it has been argued that WM and Gf share
common cognitive processes such as capacity constraints (e.g.
Fukuda, Vogel, Mayr, & Awh, 2010; Halford, Cowan, & Andrews,
2007) or the ability to resolve interference (Engle, 2002; Wiley,
Jarosz, Cushen, & Colflesh, 2011). Because of this process overlap, it has been hypothesized that improving WM skills would
facilitate performance in Gf tasks (Buschkuehl & Jaeggi, 2010;
Jaeggi, Buschkuehl, Jonides, & Perrig, 2008; Morrison & Chein,
2011). Another key feature for the success of WM training might
be the use of adaptive designs that continuously and individually
adjust the task difficulty according to the participants’ performance. The adaptive procedure allows participants to train at their
individual capacity limits and requires a continuous engagement of
executive control. The idea of adaptivity is based on the assumption that training is optimal if the task is neither too hard nor too
easy since participants will otherwise disengage due to frustration
or boredom (cf. Jaeggi et al., 2008, 2010b; Jaeggi, Buschkuehl,
Jonides, & Shah; 2012; Jaeggi, Buschkuehl, Perrig, & Meier,
2010a). The adaptivity concept is in line with Bjork and Bjork’s
(2011) concept of desirable difficulties, which promotes learning.
Almost all of the recent WM-based interventions have used adaptive techniques (e.g. Anguera et al., 2012; Buschkuehl et al., 2008;
Chein & Morrison, 2010; Jaeggi, Buschkuehl, Jonides, & Shah,
2011; Jaeggi et al., 2008, 2010b; Klingberg et al., 2005; Klingberg,
Forssberg, & Westerberg, 2002; Loosli, Buschkuehl, Perrig, &
Jaeggi, 2012; Richmond, Morrison, Chein, & Olson, 2011), and in
some of those studies, adaptive versus nonadaptive interventions
were explicitly compared, and an advantage for adaptive interventions was found (e.g., Klingberg et al., 2002, 2005; Lilienthal,
Tamez, Shelton, Myerson, & Hale, 2013). Furthermore, studies in
which an adaptive procedure was not used failed to show transfer
(Craik et al., 2007; Li et al., 2008), suggesting that adaptivity is
one of the key features for training success.
When it comes to older adults, only a few training studies have
focused on WM as a training domain. The few published studies in
which participants were trained on WM-related skills reveal an
inconsistent pattern of results ranging from null-effects to quite
substantial generalizations (Hindin & Zelinski, 2012; MelbyLervåg & Hulme, 2013; Uttal et al., 2013; Zelinski, 2009). These
inconsistencies might be the result of different training paradigms,
different training frequencies, and differences in the selected criterion tests, as well as individual differences (cf. Jaeggi et al.,
2012; Shah, Buschkuehl, Jaeggi, & Jonides, 2012). For example,

Dahlin, Neely, Larsson, Backman, and Nyberg (2008) reported
functional activation changes after WM updating training; however, in contrast to what they observed in their young adult sample,
the investigators found those activation changes were not accompanied by significant changes in behavior in their older adult
sample, except for the task on which the participants had been
trained. Buschkuehl et al. (2008) as well as Zinke, Zeintl, Eschen,
Herzog, and Kliegel (2011) observed improvements in WM in
old-old adults after WM training; however, the improvements were
restricted to tasks that were closely related to the trained domain.
A similar pattern of positive near transfer within the trained
domain of WM was reported by Richmond et al. (2011), as well as
by Li et al. (2008). On the other hand, Brehmer et al. (2011)
observed effects in older adults that went beyond mere improvements in the trained WM domain in that they found positive effects
in measures of sustained attention and episodic memory. Schmiedek, Lövdén, and Lindenberger (2010) reported improvements in
Gf in old adults after a complex cognitive intervention that consisted of training not only on WM but also on perceptual speed and
episodic memory that lasted several months. Improvements in
measures of Gf are also reported after much shorter interventions.
For example, Borella, Carretti, Riboldi, and De Beni (2010) observed improvements in Gf and other executive control tasks after
just a few days of WM training (see Carretti, Borella, Zavagnin, &
De Beni, in press, for a recent replication). Finally, there are other
reports of improvements in Gf in older adults after interventions
that target processes related to WM skills and executive control
(Basak, Boot, Voss, & Kramer, 2008; Karbach & Kray, 2009;
Tranter & Koutstaal, 2008).
The current study was based on previous research with young
adults and children, which demonstrated improvements in measures of Gf after training on an n-back task (Jaeggi et al., 2008,
2010b, 2011). An n-back task requires participants to process a
stream of stimuli (e.g., letters) and indicate for each letter whether
it is the same stimulus as the one presented n positions back in the
sequence. The difficulty of the task can be parametrically varied by
increasing the value of n (e.g. Jonides et al., 1997). N-back tasks
require participants to continuously update mental representation
of the target items while dropping irrelevant items from consideration, in addition to simple storage processes (Conway et al., 2005;
Jonides et al., 1997). In addition to having a “high face validity as
a measure of WM due to the attentional requirement to continuously update the stimuli being held on-line” (Hill et al., 2010, p. 4),
n-back tasks are usually fairly well correlated with measures of Gf
(e.g. Jaeggi et al., 2010a, 2010b; Kane, Conway, Miura, & Colflesh, 2007), suggesting at least some process overlap, which
might facilitate transfer (cf. Jaeggi et al., 2010b). As discussed
previously, it has been argued that this process overlap is most
likely related to the ability to resist distraction and resolve interference (Gray, Chabris, & Braver, 2003; Jaeggi et al., 2011; Wiley
& Jarosz, 2012).
The foremost goal of the current study was to examine the
efficacy of an adaptive computer-based WM intervention in
healthy, community-dwelling older adults. To that end, we trained
two groups of participants on a single n-back task for either 10 or
20 sessions over the course of 5 weeks. We compared the training
groups’ performance in nontrained measures of WM and visuospatial skills before and after training with the performance of a
no-contact control group. Four standardized and widely used clin-

WORKING MEMORY IN OLDER ADULTS

We used a pretest/posttest randomized controlled trial design
with two experimental groups and a control group (CG). All
participants were tested twice in an interval of 5 weeks (baseline
and posttest). The two experimental groups underwent a computerbased WM intervention between the two assessments, which differed in terms of training frequency, in that participants were asked
to train either twice or four times per week distributed over the
course of 5 weeks. The low frequency group (Ex10) trained twice
a week for a total of 7–12 sessions (mean: 9.42 sessions; SD ⫽
0.96; 200 –250 min total), whereas the high frequency group
(Ex20) trained four times a week for a total of 18 –23 sessions
(mean: 19.60 sessions; SD ⫽ 0.60; 450 –500 min total). The CG
did not train and was appointed to control for test–retest effects.
The study was approved by the Ethics Committee of Prague
Psychiatric Center (ref. no. 122/09), and informed written consent
was obtained from all participants.

or on a laptop that was provided for the duration of the study.2
Participants were required to train at home on their own, according
to their intervention-specific schedule (discussed previously). That
is, there was no contact between the participants and the research
team during the intervention period, and the training data files
were retrieved after training completion.3 The computerized training program saved the participants’ responses together with a time
stamp that allowed us to assess the training time for each session
and whether participants adhered to the required training schedule.
All participants received a nominal payment of 500 CZK after the
second assessment (i.e., about 29 USD), regardless of condition or
performance. They were given a brief report of their performance
in both assessments, and everyone received the training program
for further personal use after study completion.
One person completed the first assessment only and withdrew
from the study due to unforeseen family events. One person was
excluded from data analyses due to a MMSE score ⬍ 27 and
considerable low scores in the other tests, suggesting a global
cognitive impairment. Another person was excluded due to failure
to comply with the training schedule. Four participants were
moved from the Ex20 to the CG as technical problems prevented
them from training. One person was moved from the Ex20 to the
Ex10 as she only trained 12 times instead of the required 20
sessions. Note that this group re-allocation was performed solely
for logistical reasons and was related neither to transfer nor training performance. Thus, the final sample size was 65 (mean age:
68.1 years; SD ⫽ 2.6; 47 women); see Table 1 for the demographic
data for each group of participants. Note that the three groups did
not differ in age, education, gender distribution, initial n-back
performance, or performance in any of the manifest cognitive
baseline assessments (all Fs ⬍ 2.19).

Participants

Baseline Assessment and Transfer Measures

Sixty-eight community-dwelling young-old adults were recruited for the study. The participants were recruited through a
study-related website, journal advertisements, and flyers and posters. Participants first underwent a standardized clinical interview
that determined their eligibility for the study. Exclusion criteria
were serious medical illness that prevented unrestricted participation in the study, neurological disorders (head trauma with loss of
consciousness, epilepsy, uncorrected sensory impairment, aphasia), a current episode of a psychiatric disorder, or diagnosed or
suspected organic mental disorder (especially dementia). We also
assessed participants‘ medication use (i.e. antidepressants, hypnotics, anxiolytics, thyroid pills, or other medication), and participants
who occasionally took hypnotics (one participant) or anxiolytics
(three participants) were asked to refrain from taking the medication 3 days and 3 nights before the pre- and posttest assessments.
Finally, we screened participants’ general cognitive status with the
Mini-Mental State Examination (MMSE; Folstein, Folstein, &
McHugh, 1975; cutoff criterion ⫽ 27), and in addition, everyone
was screened for depression with the Geriatric Depression
Scale–15 (Yesavage & Sheikh, 1986; cutoff criterion ⫽ 7, scores
between 8 and 10 were followed up with a psychiatric clinical
interview).1 Once participants were determined to be eligible for
the study, they were randomly assigned to either one of the
intervention groups or the control group. Participants who were
assigned to the intervention groups trained on their own computers

In order to assess specific training effects, we administered the
same single n-back task that was used for training at the pre- and
posttest sessions (discussed later). The dependent measure was the
average n-back level obtained in those sessions.
We assessed transfer in two domains, WM and visuospatial
skills. The first target construct, WM, was represented by Digit
Span (DS) and Letter Number Sequencing (LNS); subtests of the
Wechsler Memory Scale–III (WMS–III; Wechsler, 1997b) and the
Wechsler Adult Intelligence Scale–III (WAIS–III; Wechsler,
1997a). The DS comprises two scores: the DS forward subtest
requires simple auditory WM capacity, immediate memory, and
attention span; individuals are asked to repeat an orally presented
digit sequence in forward order. The DS backward subtest assesses
more complex WM by requiring individuals to repeat an orally
presented digit sequence in backward order. The dependent variable is the sum of correctly repeated forward and backward sequences. In the LNS, a series of orally presented letters and

ical measures of WM and visuospatial skills were administered
and subsequently combined into a composite score for each domain in order to increase measurement quality (Conway et al.,
2005; Kane, et al., 2004). In addition to investigating transfer to
WM and visuospatial skills, we were especially interested to see
whether training frequency and training gain predicted the extent
of transfer in both constructs, which would thereby extend previous findings demonstrating either a dose–response effects in Gf
(Basak et al., 2008; Jaeggi et al., 2008) or a relationship between
training gain and transfer on Gf (Jaeggi et al., 2011).

Method
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

3

Design

1
We only had to follow up with one participant who had a score of 10;
however, her data were included as the clinical interview revealed that she
did not fulfill the criteria for depression.
2
All but one participant used his or her own computer for training.
3
All participants were given detailed written instructions on how to run
the program and how to do the task, along with contact information in case
they needed assistance; however, none of the participants made use of this
opportunity.

STEPANKOVA ET AL.

4
Table 1
Demographic Information

Control groupa

Experimental group
(10 sessions)b

Experimental group
(20 sessions)c

Variable

Mean (SD)

Range

Mean (SD)

Range

Mean (SD)

Range

Age (years)
Years of education

68.08 (3.01)
14.72 (2.84)

65–74
11–20

67.95 (2.19)
15.30 (3.18)

65–72
11–21

68.15 (2.62)
14.90 (3.19)

65–74
10–21

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Note. Control group: N ⫽ 25; Experimental groups: N ⫽ 20.
Gender: a 18 women and seven men. b 15 women and five men.

numbers are presented in a random order. Participants are required
to reorder and repeat the list by first reproducing the numbers in
ascending order and then the letters in alphabetical order. The
dependent variable is the number of correctly repeated lists. The
LNS requires sequencing, mental manipulation, short-term auditory memory, visual–spatial imaging, and processing-speed
(Crowe, 2000) and is regarded as a good clinical test to measure
WM functioning (Hill et al., 2010; Shelton, Elliott, Hill, Calamia,
& Gouvier, 2009).
The second target construct, visuospatial skills, was operationalized by two tasks—Block Design (BD) and Matrix Reasoning
(MR). Both are nonverbal subtests of the WAIS–III (Wechsler,
1997b). Both tests are often used as proxy for Gf (e.g., Bugg et al.,
2006; Friedman et al., 2006), and they are reported to have high g
loadings (Colom, Jung, & Haier, 2006; Roivainen, 2010). In the
MR test, participants are presented with a display of geometric
shapes or patterns, with one shape or pattern missing. They are
required to either name or point to the correct answer alternative to
complete the pattern out of five response options. The test is
thought to require visual information processing, abstract reasoning skills, learning ability, and mental flexibility (Groth-Marnat,
2009, p. 156). The BD test requires participants to organize colored blocks to match pictures on cards and employs visual perception of abstract designs, spatial processing, visuomotor coordination, and processing speed (e.g. Groth-Marnat, 2009; Kaufman,
2006, p. 401). Whereas there was no time limit on MR, BD was
timed to a maximum of 21 min as defined in the handbook of the
WAIS–III (Wechsler, 1997b). The dependent measure was determined according to the handbook of the WAIS-III (Wechsler,
1997a). In the MR, it corresponded to the number of correctly
solved items, which was also true for the BD, but in addition, the
scoring depended on the amount of time needed to solve each item.
Note that since there are no parallel-test versions of our transfer
measures, the same task versions were used in the pre- and posttest
sessions.

Training Task
We used a verbal version of the n-back task as an intervention
task, which was based on other n-back interventions used before
(cf. Jaeggi et al., 2008, 2010a, 2011). Participants were presented
with a sequence of large yellow capital letters presented in the
middle of a screen with a blue background. Participants had to
indicate whether the currently presented letter was the same as the
one presented n positions back in the sequence by pressing the
spacebar on the keyboard (no responses were required for nontargets). Presentation time was 500 ms, and the interstimulus interval

c

14 women and six men.

was 1,500 ms. Each training session consisted of 20 blocks, each
lasting for about 1 min and consisting of 20 ⫹ n stimuli. There
were six targets per block that were presented at random positions
in the sequence. Each session took about 25 min to complete. The
program continuously adapted the level of difficulty according to
each individual’s performance after each block (0 –2 errors: the
level of n was increased by one for the next block; 3–5 errors: the
level remained the same; 6 or more errors: the level of n was
decreased by one for the next test). Each session started with Level
1 (1-back). At the end of each session, participants received
feedback in that they were shown a diagram depicting the levels of
n they reached in that particular session. The dependent measure
consisted of the mean level of n reached in each training session.

Data Analyses
We assessed specific training effects with a univariate analysis
of covariance (ANCOVA) using the posttest n-back performance
as dependent variable, the pretest n-back performance as a covariate, and group (CG, Ex10, Ex20) as a between-subject factor. We
hypothesized that the two training groups would outperform the
CG and that the Ex20 group would outperform the Ex10 group. To
test those predictions, we calculated Helmert contrasts in order to
compare performance on the group level (i.e., CG vs. Ex10 and
Ex20; Ex10 vs. Ex20).
The change in cognitive performance in the transfer measures
was assessed with two composite scores. Although our pre- and
posttest task selection can be clearly assigned to WM (consisting
of DS and LNS) or visuospatial skills (consisting of MR and BD)
on a theoretical level, we nevertheless aimed to confirm this
distinctiveness statistically. To do so, we inspected the correlation
coefficients between the composite measures (cf. Table 2), and we

Table 2
Correlation Matrix (Based on the Pretest Scores)
Variable
Working memory
Digit Span
Letter–Number Sequencing
Visuospatial skills
Matrix Reasoning
Block Design

WM

VSS

DS
—
0.48ⴱⴱⴱ

LNS

0.13
0.16

0.08
0.10

MR

BD

—
0.46ⴱⴱⴱ

—

—

Note. N ⫽ 65. WM ⫽ working memory; VSS ⫽ visuospatial skills;
DS ⫽ Digital Span; LNS ⫽ Letter–Number Sequencing; MR ⫽ Matrix
Reasoning; BD ⫽ Block Design.
ⴱⴱⴱ
p ⬍ .0001.

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

WORKING MEMORY IN OLDER ADULTS

conducted a principal component analysis on the pretest data,
which clearly revealed two factors explaining 73% of the total
variance. As expected, DS and LNS loaded on one factor, and MR
and BD loaded on the other one, confirming their suitability for the
subsequent composite analyses.
Next, we divided the pretest score of each task (DS, LNS, MR,
and BD) by the standard deviation (SD) of the respective pretest
scores. We also divided the posttest scores of each task by the SD
of the respective pretest scores. This resulted in test scores that
were standardized in units of SD Note that the SD was calculated
across all three groups. Finally, the standardized scores of the WM
tasks (DS and LNS) were averaged for the pretest and for the
posttest; the same was done with the two visuospatial skills tasks
(MR and BD). These composite measures were used in the forthcoming analyses.
Transfer was assessed with univariate ANCOVAs using the
posttest composite as dependent variable, the pretest composite as
a covariate, and group (CG, Ex10, Ex20) as a between-subject
factor.4 Again, we hypothesized that the two experimental groups
would outperform the CG and that the Ex20 group would outperform the Ex10 group. Consequentially, we calculated Helmert
contrasts in order to compare transfer performance on the group
level (i.e., CG vs. Ex10 and Ex20; Ex10 vs. Ex20).
Finally, we correlated the gain in WM and visuospatial skills
with the gain in n-back performance as measured with the pre- and
posttest n-back task in order to investigate the relationship of
training improvement and extent of transfer.

Results
Specific Training Effects
The ANCOVA analyzing specific training gains as assessed
with the pre- and posttest n-back task revealed that the covariate
(pretest performance) was significantly related to the posttest per2
formance, F(1, 60) ⫽ 9.56, p ⬍ .001, ␩partial
⫽ .14. There was also
a significant effect of group after controlling for pretest perfor2
mance, F(1, 61) ⫽ 13.30, p ⬍ .001, ␩partial
⫽ .31. The participants
of the 10-day training group improved their scores by 0.97 (SD ⫽
0.85) n-back levels and the 20-day training group by 1.39 (SD ⫽ 1.22)
n-back levels. The CG retained their original scores (gain ⫽ 0.07,
SD ⫽ 0.40); see Figure 1. Planned contrasts revealed that both
experimental groups outperformed the CG in terms of n-back gain
from pretest to posttest regardless of the number of training days,
t(60) ⫽ 4.94, p ⬍ .001 (one-tailed, r ⫽ .54). Further, there was a
trend showing that the n-back improvement of the Ex20 group was
larger than the improvement of the Ex10 group, t(60) ⫽ 1.54, p ⫽
.06 (one-tailed, r ⫽ .19).

Transfer Effects
The descriptive data for all manifest pre- and posttest tasks as
well as the composite scores are reported in Table 3 as a function
of group. Note that there were no significant group differences in
any of the pretest measures, neither in the manifest nor in the
composite variables (all Fs ⬍ 2.52).
Concerning transfer on WM, the pretest performance (the covariate) was significantly related to the posttest performance, F(1,
2
61) ⫽ 108.76, p ⬍ .001, ␩partial
⫽ .64. There was also a significant

5

Figure 1. Specific training effects. Visualized are the means of the
posttest scores adjusted by the effect of the pretest covariate. Error bars
represent standard errors of the mean.

effect of training time after controlling for pretest performance,
2
⫽ .11; the adjusted group means
F(2, 61) ⫽ 3.59; p ⬍ .05, ␩partial
are visualized in Figure 2. Planned contrasts revealed that regardless of the number of training sessions, training on the n-back task
resulted in improved WM, t(61) ⫽ 2.68, p ⬍ .01 (one-tailed, r ⫽
.32) as compared with the CG. The contrast between Ex10 and
Ex20 was not significant (p ⫽ .49, one-tailed).
Regarding transfer on visuospatial skills, the covariate was
significantly related to the posttest performance, F(1, 61) ⫽
2
149.03, p ⬍ .001, ␩partial
⫽ .71). There was also a significant effect
of training time after controlling for pretest performance, F(2,
2
61) ⫽ 7.28; p ⬍ .01, ␩partial
⫽ .19, the adjusted group means are
visualized in Figure 2. Planned contrasts revealed that regardless
of the number of training sessions, training on the n-back task
resulted in improved visuospatial skills, t(61) ⫽ 3.29, p ⬍ .001
(one-tailed, r ⫽ .39) as compared with the CG. In addition, the
Ex20 group showed a significantly larger gain in visuospatial skills
than the Ex10 group, t(61) ⫽ 1.83, p ⬍ .05 (one-tailed, r ⫽ .23).
To further confirm this dose–response effect, we considered Cohen’s d (i.e., the standardized effect size; cf. Table 3). The lowest
effect size was observed in the control group (d ⫽ 0.34), followed
by the effect size of the Ex10 group (d ⫽ 0.60), and the effect size
of the Ex20 group (d ⫽ 0.98). In order to estimate the impact of
training frequency, we subtracted the effect size of the control
group from each of the experimental groups, resulting in d ⫽ 0.26
for the Ex10 group and d ⫽ 0.64 for the Ex20 group, which further
supports our notion of a dose–response effect in the visuospatial
domain.
Finally, we used Spearman correlations to examine the relationship between n-back performance gain as measured with the
n-back task assessed in the pre- and posttest sessions and the gains
4
Note that an alternative way to analyze the data would be through a
repeated-measures analysis of variance (ANOVA); however, an ANCOVA
is more powerful than a repeated-measures ANOVA because the covariate
explains additional variance (cf. Weinfurt, 2002). Nevertheless, we also
calculated repeated-measures ANOVAs and the resulting significance levels were identical to the ones resulting from the ANCOVAs.

STEPANKOVA ET AL.

6

Table 3
Descriptive Data For All Pre- and Posttest Measures on a Manifest and Construct Level as Function of Group
Pretest

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Variable
Experimental group (10 sessions)
n-back
Digit Span
Letter–Number Sequencing
Block Design
Matrix Reasoning
Working memory (composite)
Visuospatial skills (composite)
Experimental group (20 sessions)
n-back
Digit Span
Letter–Number Sequencing
Block Design
Matrix Reasoning
Working memory (composite)
Visuospatial skills (composite)
No-contact control group
n-back
Digit Span
Letter–Number Sequencing
Block Design
Matrix Reasoning
Working memory (composite)
Visuospatial skills (composite)

Posttest

N

Mean

SD

Min

Max

Mean

SD

Min

Max

p

20
20
20
20
20
20
20

2.4
18.65
9.5
42.25
18.15
4.86
4.36

0.36
3.7
1.5
8.89
4.23
0.79
0.77

1.75
12
7
25
10
3.35
2.61

2.9
24
12
58
23
6.20
5.32

3.38
20.25
10.4
46.2
20.25
5.30
4.81

0.92
4.01
2.26
8.84
3.77
1.01
0.73

2.4
12
7
25
10
3.35
3.04

5.8
28
15
64
25
7.48
5.71

ⴱⴱⴱ

20
20
20
20
20
20
20

2.49
18.95
9.75
38.35
17.5
4.97
4.06

0.4
3.49
2.49
8.68
4.45
0.91
0.84

1.9
14
4
22
8
3.15
2.34

3.35
27
14
51
24
7.10
5.15

3.88
20.5
10.6
44.9
21.1
5.38
4.83

1.29
3.53
2.62
8.77
2.95
1.01
0.72

2.65
14
5
31
15
3.53
3.38

7.4
29
14
57
25
7.37
5.93

ⴱⴱⴱ

25
25
25
25
25
25
25

2.35
17.44
9.96
36.92
15.76
4.81
3.80

0.41
3.86
2.03
8.46
4.93
0.90
0.87

1.5
10
7
24
4
3.33
1.79

3.15
25
14
55
23
6.32
4.95

2.43
18.4
9.64
39.84
17.04
4.86
4.10

0.47
3.8
1.78
9.52
5.02
0.84
0.92

1.7
13
7
23
6
3.48
2.12

3.6
24
12
57
24
6.20
5.72

ⴱⴱ
ⴱ
ⴱⴱ
ⴱⴱ
ⴱⴱⴱ
ⴱⴱⴱ

ⴱ

0.09
ⴱⴱⴱ
ⴱⴱ
ⴱ
ⴱⴱⴱ

0.38
0.09
0.32
ⴱ
ⴱ

0.67
ⴱⴱ

Effect size
(Cohen’s d)
1.39
0.41
0.47
0.45
0.52
0.49
0.60
1.46
0.44
0.33
0.75
0.95
0.43
0.98
0.18
0.25
⫺0.17
0.32
0.26
0.06
0.34

Note. In order to calculate the reported statistics for the composite measures, we divided the pretest values of the manifest variables by the standard
deviation from the pretest session (based on all three groups) and then averaged in each domain (i.e., Digit Span and Letter–Number Sequencing for working
memory and Block Design and Matrix Reasoning for visuospatial skills). The same procedure was used for the posttest statistics; p values are based on
paired t tests (posttest vs. pretest). Min ⫽ minimum; max ⫽ maximum.
ⴱ
p ⬍ .05. ⴱⴱ p ⬍ .01. ⴱⴱⴱ p ⬍ .001.

in WM and visuospatial skills. We found that the improvement in
n-back performance was positively related to the improvement in
both, WM (Spearman ␳ ⫽ 0.32; p ⬍ .05) and visuospatial skills
(Spearman ␳ ⫽ 0.26; p ⬍ .05).

Figure 2. Transfer effects. Visualized are the means of the posttest scores
adjusted by the effect of the pretest covariate for both composite scores.
Error bars represent standard errors of the mean. Planned contrasts revealed
that both experimental groups outperformed the control group in working
memory and visuospatial skills, and further, that the 20-day training group
outperformed the 10-day training group in the visuospatial composite.

Discussion
The present study revealed that older adults can significantly
improve performance in an n-back task as a result of training for
a period of either 10 or 20 sessions. But more important, we
observed reliable transfer to nontrained composite measures of
WM and visuospatial skills, thereby replicating but also extending
earlier work with children and young adults (Jaeggi et al., 2008,
2010a, 2010b, 2011). Additionally, we found a positive relationship between training gain and transfer in both criterion measures,
suggesting that the intervention was the contributing factor to
transfer (cf. Chein & Morrison, 2010; Jaeggi et al., 2011; von
Bastian & Oberauer, 2013; Zhao, Wang, Liu, & Zhou, 2011).
Finally, higher training frequency resulted in a larger gain in
visuospatial skills, which is consistent with previous findings as
well (Basak et al., 2008; Dahlin, Bäckman, Neely, & Nyberg,
2009; Jaeggi et al., 2008).
As pointed out earlier, it has been difficult to demonstrate
evidence for transfer that goes beyond tasks that are closely related
to the training task in older adults (e.g. Buschkuehl et al., 2008;
Dahlin et al., 2008; Li et al., 2008; Zinke et al., 2011). Nevertheless, the present results add to the evidence for the malleability of
visuospatial skills (Uttal et al., 2013) and, more specifically, to the
few studies reporting transfer on visuospatial skills following WM
training in older adults. But our data also extend the previous
literature in important ways. For example, Schmiedek, Lövdén, &
Lindenberger (2010) trained their participants on a complex train-

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

WORKING MEMORY IN OLDER ADULTS

ing regimen that consisted of episodic memory, processing speed,
and WM for a period of several months. Although Schmiedek et al.
found Gf improvements on one manifest variable in their older
adult population, that is, in the Raven’s Advanced Progressive
Matrices (APM), the improvement was not significant on a latent
level, which included three subtests of the Berlin Intelligence
Structure Test (BIS; Jäger, Süss, & Beauducel, 1997) in addition to
the APM.5 The authors hypothesized that the timed nature of most
of the Gf tasks used worked against transfer in these tasks (cf. also
Hofland, Willis, & Baltes, 1981). This hypothesis is supported in
so far as their older participants showed transfer on two tasks with
less time pressure, one of them being the APM. In our study, we
employed tests with less obvious (BD) or no (MR) time limit,
which could have been a reason why we observed transfer. However, the studies of Borella et al. (2010) and Carretti et al. (in press)
do not seem to support this notion, as these authors observed
increased performance in a timed Gf measure, the Cattell Culture
Fair Test of Intelligence (CFT; Cattell & Cattell, 1963). One has to
keep in mind, though, that these three training studies differed
considerably in their methodology, especially when it comes to the
employed training paradigm (training on tasks from different domains by Schmiedek et al. vs. training on a single variant of a
complex span task by Borella et al., 2010, and Carretti et al., in
press) but also training time (⬎ 100 days by Schmiedek et al.
versus 3 days by Borella et al. and Carretti et al.). Given these and
other differences between our study and the ones discussed previously, it is impossible to conclusively determine whether the
timing of Gf tasks is important for transfer effects to occur.
Nevertheless, our results, as well as the results from Schmiedek et
al., Borella et al., and Carretti et al., are consistent with evaluations
by Schaie & Willis (1986) and Lövdén, Bäckman, Lindenberger,
Schafer, and Schmiedek (2010) in which they concluded that older
adults are indeed capable of improving higher cognitive processes
and that there are certain cognitive interventions that seem to be
able to induce plasticity in old age.
Moody (2009) argued that the transfer effects on Gf as reported
in Jaeggi et al. (2008) were most likely the result of the visuospatial nature of the dual n-back task that was used for training.
Although this argument does not consider the fact that the dual
n-back training task entailed a verbal component as well, one could
still speculate that the visuospatial part was the main contributor to
the transfer effects. The present study now provides evidence that
the visuospatial nature of the task is not driving the transfer to
visuospatial skills since the transfer emerged by training on a task
that was clearly nonvisuospatial, suggesting that the mechanisms
underlying n-back training and transfer do not rely on materialspecific processes (see also Borella et al., 2010; Carretti et al., in
press).
The observed dose-response effect, together with the positive
correlation of n-back training gain and the improvement in both
criterion measures, provides strong evidence that the transfer effects indeed resulted from the intervention and not from training
unrelated factors such as an expectancy bias (see also Basak et al.,
2008; Chein & Morrison, 2010; Jaeggi et al., 2008, 2011). But
what could be the common underlying mechanism that drives both,
the improvement in WM as well as the improvement in visuospatial skills? It has been argued before that process overlap between
the training and transfer task is an essential prerequisite for transfer
(Dahlin et al., 2008; Jaeggi et al., 2010b; Lustig et al., 2009). Since

7

we observed transfer in both domains, it is likely that both domains
share processes with the training task, and we speculate that
specific WM-related processes such as storage and manipulation
are the most likely candidates since they are necessary to perform
well in both domains (e.g. storing and updating the order of the
letters and numbers in WM during LNS, or holding and comparing
multiple solution principles in WM during MR).
The question is now why there was a dose–response effect in
visuospatial skills, but not in WM. Interestingly, a similar pattern
was observed in a previous study in young adults, that is, a
dose–response effect in Gf after training on WM skills, as well as
a WM effect irrespective of dose (Jaeggi et al., 2008). One possible
explanation for this result pattern is that the WM effects might be
easier to obtain due to the domain specificity, that is, due to the
fact that there is a stronger process overlap between the trained
task and the WM tasks; thus, the improvements in WM occur
regardless of training frequency. Alternatively, older adults might
reach their WM capacity limit fairly quickly (Reuter-Lorenz &
Cappell, 2008; Schneider-Garces et al., 2010), that is, within 10
days of training. In contrast, there might be more room to improve
visuospatial skills because performance in such tasks relies on
WM processes in addition to other processes that do not seem to
reach a capacity limit so quickly. In the cognitive domain, such
processes could be an improved ability to suppress distraction,
attentional control, speed of processing, or strategy changes (Jaeggi et al., 2008; 2010b; Morrison & Chein, 2011). In addition,
noncognitive processes might come into play as well, such as
increased confidence, self-efficacy, reduced anxiety, and persistence to stick with the task and not giving up regardless of the
difficulty of the task, which is certainly a requirement to perform
well in the adaptive n-back training task but also in the reasoning
measures (e.g. Borella et al., 2010; Carretti et al., in press; Chein
& Morrison, 2010; Duckworth, Peterson, Matthews, & Kelly,
2007; Hayslip, 1989; Hayslip, Maloy, & Kohl, 1995; Jaeggi et al.,
2011).
Despite the clear results, we acknowledge several limitations
with our current study. First is the fact that our participants were
relatively young (age range: 65–75 years), which could have
facilitated the occurrence of transfer (Buschkuehl et al., 2008;
Singer, Lindenberger, & Baltes, 2003; Zinke et al., 2011). Further,
we found that our population performed on a relatively high level
from the start, which might be regarded as detrimental for the
generalizability of the current data. The high pretest scores of our
participants might be indicative of a high education level (Salthouse, 1993), and indeed, our sample had higher levels of education than can be found in the Czech general population in this age
group (p ⬍ .001), suggesting a self-selection bias towards highly
functioning participants. However, this seems to be a common
phenomenon with intervention studies in old age (Lezak, Howieson, & Loring, 2004, p. 296). On the other hand, such high initial
ability levels could also imply that there may not be as much room
for improvement as there could be in lower functioning participants, thus making our findings even more impressive (cf. Zinke et
al., 2011). Yet, it might be of interest for future researchers to
investigate whether participants with lower education and accom5
Note that the improvement on Gf on the latent level was significant in
their young adult sample.

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

8

STEPANKOVA ET AL.

panying lower pretest scores would obtain even larger gains than
those in our well-educated group. Such a prediction is not without
controversy though. Previous research has found that education
level is not necessarily indicative for training outcome. For example, it has been shown that gains in speed of processing did not
correlate with education (Ball, Edwards, & Ross, 2007) and education did not affect training-related gains in memory either (Stigsdotter Neely & Bäckman, 1995). Verhaeghen & Marcoen (1996)
even reported that individuals with high initial WM capacity show
the largest gains in episodic memory as a result of strategy training. It is important to note at this point that although our sample
was performing very well already in the pretest, this did not
preclude them from showing transfer because there was enough
room for improvement in all four manifest tasks. Unfortunately,
we do not know how long our transfer effects last due to a lack of
follow-up assessment; however, there are previous reports that
show lasting effects of training in children (e.g. Jaeggi, et al.,
2011) as well as in older adults (Borella et al., 2010; Carretti et al.,
in press), suggesting that such effects might be retained (but see
Buschkuehl et al., 2008). Nevertheless, it seems reasonable to
predict that in order to obtain maximal retention, regular booster
sessions might be necessary, especially in older adults (Ball et al.,
2007; Lövdén et al., 2010). Finally, the lack of an active control
group can be seen as a limiting factor. It has been argued that
no-contact control groups do not control for the motivation of
participants, which may lead to suboptimal performance and may
wrongly magnify an intervention effect (Shipstead, Redick, &
Engle, 2012); however, it has also been argued that no-contact
control groups are adequate to control for test-retest effects (Chein
& Morrison, 2010; see also Silverman & Hinshaw, 2008), and
furthermore, studies that have used both placebo and no-contact
control groups did not observe any significant group differences in
motivation (Bergman Nutley et al., 2011). In our case, the nocontact control group performed as well as the experimental
groups at pretest and also showed healthy test–retest effects.
Therefore, we assume that even participants in the control group
tried hard to achieve their best performance in the pre- and posttest
sessions. Furthermore, the fact that we used an unsupervised
training procedure rules out an unspecific experimenter effect that
could indeed work against our hypothesis, as there is literature
showing that (supervised) group sessions (i.e., socializing) can
have a positive impact on cognitive performance (Verhaeghen et
al., 1992; Ybarra et al., 2008). Thus, the fact that we actually did
observe improvement in training (along with transfer effects) suggests that our effects go beyond unspecific effects of experimenter
presence or socializing. In addition, as we pointed out earlier, the
dose–response relationship as observed in visuospatial ability, as
well as the positive relationship between training gain and transfer
in general, suggests that the effects were most likely a specific
result of the intervention. Finally, we would like to acknowledge
that we used the same test versions in the pre- and posttest
sessions. Unfortunately, there are no parallel-test versions available for the standardized clinical tests that we used here, and rather
than splitting the tests in half and thereby reducing their excellent
psychometric characteristics, we decided to use the same tests for
both test occasions, a procedure that has been used by others
before (cf. also Jobe et al., 2001; Klingberg et al., 2005; Rueda,
2005; Schellenberg, 2004; Schmiedek, Lövdén, & Lindenberger,
2010; von Bastian & Oberauer, 2013).

To conclude, our data demonstrate generalizing effects to composite scores reflecting WM and visuospatial skills in young-old
healthy adults after a verbal n-back intervention. Furthermore, we
observed a dose–response effect in that those participants who
completed more training sessions showed more performance gain
on visuospatial skills. Our work adds to the accumulating evidence
for transfer effects in old adults by means of an easily accessible
noncommercial computer-based program that can be used independently at home. Such unsupervised interventions are very important from an applied point of view and are very economical as
they are cost-effective and only require a minimal amount of
personnel for administration. Future projects may aim to further
investigate the breadth of transfer by including a broader variety of
tests, which would help to identify domain-specific and domaingeneral factors. Further, it is important to validate the reported
effects on larger and more heterogeneous samples by including
participants with a wider age and ability range in order to make
further claims about the generalizability of WM interventions.

References
Alzheimer’s Association. (2011). 2011 Alzheimer’s disease facts and figures. Alzheimer’s & Dementia, 7, 208 –244. doi:10.1016/S15525260(11)00040-9
Anguera, J. A., Bernard, J. A., Jaeggi, S. M., Buschkuehl, M., Benson,
B. L., Jennett, S., . . . Seidler, R. D. (2012). The effects of working
memory resource depletion and training on sensorimotor adaptation.
Behavioural Brain Research, 228, 107–115. doi:10.1016/j.bbr.2011.11
.040
Ball, K., Berch, D. B., Helmers, K. F., Jobe, J. B., Leveck, M. D.,
Marsiske, M., . . . Willis, S. L. (2002). Effects of cognitive training
interventions with older adults: A randomized controlled trial. JAMA:
The Journal of the American Medical Association, 288, 2271–2281.
doi:10.1001/jama.288.18.2271
Ball, K., Edwards, J. D., & Ross, L. A. (2007). The impact of speed of
processing training on cognitive and everyday functions. Journals of
Gerontology, Series B: Psychological Sciences and Social Sciences,
62(Special Issue 1), 19 –31.
Baltes, P. B., & Lindenberger, U. (1997). Emergence of a powerful
connection between sensory and cognitive functions across the adult life
span: A new window to the study of cognitive aging? Psychology and
Aging, 12, 12–21. doi:10.1037/0882-7974.12.1.12
Basak, C., Boot, W. R., Voss, M. W., & Kramer, A. F. (2008). Can training
in a real-time strategy video game attenuate cognitive decline in older
adults? Psychology and Aging, 23, 765–777. doi:10.1037/a0013494
Bergman Nutley, S., Söderqvist, S., Bryde, S., Thorell, L. B., Humphreys,
K., & Klingberg, T. (2011). Gains in fluid intelligence after training
non-verbal reasoning in 4-year-old children: A controlled, randomized
study. Developmental Science, 14, 591– 601. doi:10.1111/j.1467-7687
.2010.01022.x
Bjork, E. L., & Bjork, R. (2011). Making things hard on yourself, but in a
good way: Creating desirable difficulties to enhance learning. In M. A.
Gernsbacher, R. W. Pew, L. M. Hough, & J. R. Pomerantz (Eds.),
Psychology and the real world: Essays illustrating fundamental contributions to society (pp. 56 – 64). New York, NY: Worth.
Borella, E., Carretti, B., Riboldi, F., & De Beni, R. (2010). Working
memory training in older adults: Evidence of transfer and maintenance
effects. Psychology and Aging, 25, 767–778. doi:10.1037/a0020683
Brehmer, Y., Rieckmann, A., Bellander, M., Westerberg, H., Fischer, H.,
& Bäckman, L. (2011). Neural correlates of training-related workingmemory gains in old age. NeuroImage, 58, 1110 –1120. doi:10.1016/j
.neuroimage.2011.06.079

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

WORKING MEMORY IN OLDER ADULTS
Bugg, J. M., Zook, N. A., DeLosh, E. L., Davalos, D. B., & Davis, H. P.
(2006). Age differences in fluid intelligence: Contributions of general
slowing and frontal decline. Brain and Cognition, 62, 9 –16. doi:
10.1016/j.bandc.2006.02.006
Buschkuehl, M., & Jaeggi, S. M. (2010). Improving intelligence: A literature review. Swiss Medical Weekly, 140, 266 –272.
Buschkuehl, M., Jaeggi, S. M., Hutchison, S., Perrig-Chiello, P., Däpp, C.,
Müller, M., . . . Perrig, W. J. (2008). Impact of working memory training
on memory performance in old-old adults. Psychology and Aging, 23,
743–753. doi:10.1037/a0014342
Carretti, B., Borella, E., Zavagnin, M., & De Beni, R. (in press). Gains in
language comprehension relating to working memory training in healthy
elderly adults. International Journal of Geriatric Psychiatry.
Cattell, R. (1971). Abilities: Their structure, growth, and action. Boston,
MA: Houghton Mifflin.
Cattell, R. B., & Cattell, H. E. P. (1963). Measuring intelligence with the
Culture Fair Tests. Champaign, IL: Institute for Personality and Ability
Testing.
Chein, J. M., & Morrison, A. B. (2010). Expanding the mind’s workspace:
Training and transfer effects with a complex working memory span task.
Psychonomic Bulletin & Review, 17, 193–199. doi:10.3758/PBR.17.2
.193
Colom, R., Jung, R. E., & Haier, R. J. (2006). Distributed brain sites for the
g-factor of intelligence. NeuroImage, 31, 1359 –1365. doi:10.1016/j
.neuroimage.2006.01.006
Conway, A. R. A., Kane, M. J., Bunting, M. F., Hambrick, D. Z., Wilhelm,
O., & Engle, R. W. (2005). Working memory span tasks: A methodological review and user’s guide. Psychonomic Bulletin & Review, 12,
769 –786. doi:10.3758/BF03196772
Cowan, N., Elliott, E. M., Scott Saults, J., Morey, C. C., Mattox, S.,
Hismjatullina, A., & Conway, A. R. A. (2005). On the capacity of
attention: Its estimation and its role in working memory and cognitive
aptitudes. Cognitive Psychology, 51, 42–100. doi:10.1016/j.cogpsych
.2004.12.001
Craik, F. I. M., Winocur, G., Palmer, H., Binns, M. A., Edwards, M.,
Bridges, K., . . . Stuss, D. T. (2007). Cognitive rehabilitation in the
elderly: Effects on memory. Journal of the International Neuropsychological Society, 13, 132–142. doi:10.1017/S1355617707070166
Crowe, S. F. (2000). Does the letter number sequencing task measure
anything more than digit span? Assessment, 7, 113–117. doi:10.1177/
107319110000700202
Dahlin, E., Bäckman, L., Neely, A. S., & Nyberg, L. (2009). Training of
the executive component of working memory: Subcortical areas mediate
transfer effects. Restorative Neurology and neuroscience, 27, 405– 419.
doi:10.3233/RNN-2009-0492
Dahlin, E., Neely, A. S., Larsson, A., Backman, L., & Nyberg, L. (2008,
June 13). Transfer of learning after updating training mediated by the
striatum. Science, 320, 1510 –1512. doi:10.1126/science.1155466
Duckworth, A. L., Peterson, C., Matthews, M. D., & Kelly, D. R. (2007).
Grit: Perseverance and passion for long-term goals. Journal of Personality and Social Psychology, 92, 1087–1101. doi:10.1037/0022-3514.92
.6.1087
Engle, R. W. (2002). Working memory capacity as executive attention.
Current Directions in Psychological Science, 11, 19 –23. doi:10.1111/
1467-8721.00160
Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). “Mini-mental
state.” A practical method for grading the cognitive state of patients for
the clinician. Journal of Psychiatric Research, 12, 189 –198. doi:
10.1016/0022-3956(75)90026-6
Friedman, N. P., Miyake, A., Corley, R. P., Young, S. E., Defries, J. C., &
Hewitt, J. K. (2006). Not all executive functions are related to intelligence. Psychological Science, 17, 172–179. doi:10.1111/j.1467-9280
.2006.01681.x

9

Fukuda, K., Vogel, E., Mayr, U., & Awh, E. (2010). Quantity, not quality:
The relationship between fluid intelligence and working memory capacity. Psychonomic Bulletin & Review, 17, 673– 679. doi:10.3758/17.5.673
Gottfredson, L. S. (1997). Why g matters: The complexity of everyday life.
Intelligence, 24, 79 –132. doi:10.1016/S0160-2896(97)90014-3
Gray, J. R., Chabris, C. F., & Braver, T. S. (2003). Neural mechanisms of
general fluid intelligence. Nature Neuroscience, 6, 316 –322. doi:
10.1038/nn1014
Groth-Marnat, G. (2009). Handbook of psychological assessment (5th ed.).
Hoboken, NJ: Wiley.
Halford, G. S., Cowan, N., & Andrews, G. (2007). Separating cognitive
capacity from knowledge: A new hypothesis. Trends in Cognitive Sciences, 11, 236 –242. doi:10.1016/j.tics.2007.04.001
Hayslip, B. (1989). Fluid ability training with aged people: A past with a
future? Educational Gerontology, 15, 573–595. doi:10.1080/
0380127890150603
Hayslip, B., Maloy, R. M., & Kohl, R. (1995). Long-term efficacy of fluid
ability interventions with older adults. Journals of Gerontology, Series
B: Psychological Sciences, 50, P141–P149. doi:10.1093/geronb/50B.3
.P141
Hill, B. D., Elliott, E. M., Shelton, J. T., Pella, R. D., O’Jile, J. R., &
Gouvier, W. D. (2010). Can we improve the clinical assessment of
working memory? An evaluation of the Wechsler Adult Intelligence
Scale–Third Edition using a working memory criterion construct. Journal of Clinical and Experimental Neuropsychology, 32, 315–323. doi:
10.1080/13803390903032529
Hindin, S. B., & Zelinski, E. M. (2012). Extended practice and aerobic
exercise interventions benefit untrained cognitive outcomes in older
adults: A meta-analysis. Journal of the American Geriatrics Society, 60,
136 –141. doi:10.1111/j.1532-5415.2011.03761.x
Hofland, B. F., Willis, S. L., & Baltes, P. B. (1981). Fluid intelligence
performance in the elderly: Intraindividual variability and conditions of
assessment. Journal of Educational Psychology, 73, 573–586. doi:
10.1037/0022-0663.73.4.573
Jaeggi, S. M., Buschkuehl, M., Jonides, J., & Perrig, W. J. (2008). Improving fluid intelligence with training on working memory. Proceedings of the National Academy of Sciences, USA, 105, 6829 – 6833.
doi:10.1073/pnas.0801268105
Jaeggi, S. M., Buschkuehl, M., Jonides, J., & Shah, P. (2011). Short- and
long-term benefits of cognitive training. Proceedings of the National
Academy of Sciences, USA, 108, 10081–10086. doi:10.1073/pnas
.1103228108
Jaeggi, S. M., Buschkuehl, M., Jonides, J., & Shah, P. (2012). Cogmed and
working memory training: Current challenges and the search for underlying mechanisms. Journal of Applied Research in Memory and Cognition, 1, 211–213. doi:10.1016/j.jarmac.2012.07.002
Jaeggi, S. M., Buschkuehl, M., Perrig, W. J., & Meier, B. (2010a). The
concurrent validity of the n-back task as a working memory measure.
Memory, 18, 394 – 412. doi:10.1080/09658211003702171
Jaeggi, S. M., Buschkuehl, M., Shah, P., & Jonides, J. (2013). The role of
individual differences in cognitive training and transfer. Memory &
Cognition. Advance online publication. doi:10.3758/s13421-013-0364-z
Jaeggi, S. M., Studer-Luethi, B., Buschkuehl, M., Su, Y.-F., Jonides, J., &
Perrig, W. J. (2010b). The relationship between n-back performance and
matrix reasoning: Implications for training and transfer. Intelligence, 38,
625– 635. doi:10.1016/j.intell.2010.09.001
Jäger, A. O., Süss, H.-M., & Beauducel, A. (1997). Berliner Intelligenzstruktur Test (BIS Test. Form 4): Handanweisung [Berlin Intelligence
Structure Test (BIS Test, Form 4): Manual]. Göttingen, Germany:
Hogrefe.
Jobe, J. B., Smith, D. M., Ball, K., Tennstedt, S. L., Marsiske, M., Willis,
S. L., . . . Kleinman, K. (2001). ACTIVE: A cognitive intervention trial
to promote independence in older adults. Controlled Clinical Trials, 22,
453– 479. doi:10.1016/S0197-2456(01)00139-8

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

10

STEPANKOVA ET AL.

Jonides, J., Lewis, R. L., Nee, D. E., Lustig, C. A., Berman, M. G., &
Moore, K. S. (2008). The mind and brain of short-term memory. Annual
Review of Psychology, 59, 193–224. doi:10.1146/annurev.psych.59
.103006.093615
Jonides, J., Schumacher, E. H., Smith, E. E., Lauber, E. J., Awh, E.,
Minoshima, S., & Koeppe, R. A. (1997). Verbal working memory load
affects regional brain activation as measured by PET. Journal of Cognitive Neuroscience, 9, 462– 475. doi:10.1162/jocn.1997.9.4.462
Kane, M. J., Conway, A. R. A., Miura, T. K., & Colflesh, G. J. H. (2007).
Working memory, attention control, and the n-back task: A question of
construct validity. Journal of Experimental Psychology Learning, Memory, and Cognition, 33, 615– 622. doi:10.1037/0278-7393.33.3.615
Kane, M. J., Hambrick, D. Z., Tuholski, S. W., Wilhelm, O., Payne, T. W.,
& Engle, R. W. (2004). The generality of working memory capacity: A
latent-variable approach to verbal and visuospatial memory span and
reasoning. Journal of Experimental Psychology: General, 133, 189 –
217. doi:10.1037/0096-3445.133.2.189
Karbach, J., & Kray, J. (2009). How useful is executive control training?
Age differences in near and far transfer of task-switching training.
Developmental Science, 12, 978 –990. doi:10.1111/j.1467-7687.2009
.00846.x
Kaufman, A., & Lichtenberger, E. O. (2006). Assessing adolescent and
adult intelligence (3rd ed..). Hoboken NJ: Wiley.
Klingberg, T., Fernell, E., Olesen, P. J., Johnson, M., Gustafsson, P.,
Dahlström, K., . . . Westerberg, H. (2005). Computerized training of
working memory in children with ADHD: A randomized, controlled
trial. Journal of the American Academy of Child & Adolescent Psychiatry, 44, 177–186. doi:10.1097/00004583-200502000-00010
Klingberg, T., Forssberg, H., & Westerberg, H. (2002). Training of working memory in children with ADHD. Journal of Clinical and Experimental Neuropsychology, 24, 781–791. doi:10.1076/jcen.24.6.781.8395
Lezak, M. D., Howieson, D. B., & Loring, D. W. (Ed.). (2004). Neuropsychological assessment (4th ed..). New York, NY: Oxford University
Press.
Li, S.-C., Schmiedek, F., Huxhold, O., Röcke, C., Smith, J., & Lindenberger, U. (2008). Working memory plasticity in old age: Practice gain,
transfer, and maintenance. Psychology and Aging, 23, 731–742. doi:
10.1037/a0014343
Lilienthal, L., Tamez, E., Shelton, J. T., Myerson, J., & Hale, S. (2013).
Dual n-back training increases the capacity of the focus of attention.
Psychonomic Bulletin & Review, 20, 135–141. doi:10.3758/s13423-0120335-6
Loosli, S. V., Buschkuehl, M., Perrig, W. J., & Jaeggi, S. M. (2012).
Working memory training improves reading processes in typically developing children. Child Neuropsychology, 18, 62–70. doi:10.1080/
09297049.2011.575772
Lövdén, M., Bäckman, L., Lindenberger, U., Schaefer, S., & Schmiedek, F.
(2010). A theoretical framework for the study of adult cognitive plasticity. Psychological Bulletin, 136, 659 – 676. doi:10.1037/a0020080
Lustig, C., Shah, P., Seidler, R., & Reuter-Lorenz, P. A. (2009). Aging,
training, and the brain: A review and future directions. Neuropsychology
Review, 19, 504 –522. doi:10.1007/s11065-009-9119-9
Melby-Lervåg, M., & Hulme, C. (2013). Is working memory training
effective? A meta-analytic review. Developmental Psychology. doi:
10.1037/a0028228
Moody, D. E. (2009). Can intelligence be increased by training on a task
of working memory? Intelligence, 37, 327–328. doi:10.1016/j.intell
.2009.04.005
Morrison, A. B., & Chein, J. M. (2011). Does working memory training
work? The promise and challenges of enhancing cognition by training
working memory. Psychonomic Bulletin & Review, 18, 46 – 60. doi:
10.3758/s13423-010-0034-0
Oberauer, K., Schulze, R., Wilhelm, O., & Süss, H.-M. (2005). Working
memory and intelligence: Their correlation and their relation: Comment

on Ackerman, Beier, and Boyle (2005). Psychological Bulletin, 131,
61– 65; author reply 72–75. doi:10.1037/0033-2909.131.1.61
Park, D. C., Lautenschlager, G., Hedden, T., Davidson, N. S., Smith, A. D.,
& Smith, P. K. (2002). Models of visuospatial and verbal memory across
the adult life span. Psychology and Aging, 17, 299 –320. doi:10.1037/
0882-7974.17.2.299
Reuter-Lorenz, P. A., & Cappell, K. A. (2008). Neurocognitive aging and
the compensation hypothesis. Current Directions in Psychological Science, 17, 177–182. doi:10.1111/j.1467-8721.2008.00570.x
Richmond, L. L., Morrison, A. B., Chein, J. M., & Olson, I. R. (2011).
Working memory training and transfer in older adults. Psychology and
Aging, 26, 813– 822. doi:10.1037/a0023631
Roivainen, E. (2010). European and American WAIS III norms: Crossnational differences in performance subtest scores. Intelligence, 38,
187–192. doi:10.1016/j.intell.2009.10.001
Rowley, J., & Slack, F. (2008). Conceptions of wisdom. Journal of
Information Science, 35, 110 –119. doi:10.1177/0165551508092269
Rueda, M. R. (2005). Training, maturation, and genetic influences on the
development of executive attention. Proceedings of the National Academy of Sciences, USA, 102, 14931–14936. doi:10.1073/pnas
.0506897102
Ryan, J. J., Sattler, J. M., & Lopez, S. J. (2000). Age effects on Wechsler
Adult Intelligence Scale–III subtests. Archives of Clinical Neuropsychology, 15, 311–317.
Salthouse, T. A. (1993). Speed mediation of adult age differences in
cognition. Developmental Psychology, 29, 722–738. doi:10.1037/00121649.29.4.722
Salthouse, T. A. (1996). The processing-speed theory of adult age differences in cognition. Psychological Review, 103, 403– 428. doi:10.1037/
0033-295X.103.3.403
Schaie, W. K., & Willis, S. L. (1986). Can decline in adult intellectual
functioning be reversed? Developmental Psychology, 22, 223–232. doi:
10.1037/0012-1649.22.2.223
Schellenberg, E. G. (2004). Music lessons enhance IQ. Psychological
Science, 15, 511–514. doi:10.1111/j.0956-7976.2004.00711.x
Schmiedek, F., Lövdén, M., & Lindenberger, U. (2010). Hundred days of
cognitive training enhance broad cognitive abilities in adulthood: Findings from the COGITO Study. Frontiers in Aging Neuroscience, 2, 27.
doi:10.3389/fnagi.2010.00027
Schneider-Garces, N. J., Gordon, B. A., Brumback-Peltz, C. R., Shin, E.,
Lee, Y., Sutton, B. P., . . . Fabiani, M. (2010). Span, CRUNCH, and
beyond: Working memory capacity and the aging brain. Journal of
Cognitive Neuroscience, 22, 655– 669. doi:10.1162/jocn.2009.21230
Shah, P., Buschkuehl, M., Jaeggi, S., & Jonides, J. (2012). Cognitive
training for ADHD: The importance of individual differences. Journal of
Applied Research in Memory and Cognition, 1, 204 –205. doi:10.1016/
j.jarmac.2012.07.001
Shelton, J. T., Elliott, E. M., Hill, B. D., Calamia, M. R., & Gouvier, W. D.
(2009). A comparison of laboratory and clinical working memory tests
and their prediction of fluid intelligence. Intelligence, 37, 283–293.
doi:10.1016/j.intell.2008.11.005
Shipstead, Z., Redick, T. S., & Engle, R. W. (2012). Is working memory
training effective? Psychological Bulletin, 138, 628 – 654. doi:10.1037/
a0027473
Silverman, W. K., & Hinshaw, S. P. (2008). The second special issue on
evidence-based psychosocial treatments for children and adolescents: A
10-year update. Journal of Clinical Child & Adolescent Psychology, 37,
1–7. doi:10.1080/15374410701817725
Singer, T., Lindenberger, U., & Baltes, P. B. (2003). Plasticity of memory
for new learning in very old age: A story of major loss? Psychology and
Aging, 18, 306 –317. doi:10.1037/0882-7974.18.2.306
Stigsdotter Neely, A., & Bäckman, L. (1995). Effects of multifactorial
memory training in old age: Generalizability across tasks and individu-

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

WORKING MEMORY IN OLDER ADULTS
als. Journals of Gerontology, Series B: Psychological Sciences and
Social Sciences, 50, P134 –P140. doi:10.1093/geronb/50B.3.P134
Tranter, L. J., & Koutstaal, W. (2008). Age and flexible thinking: An
experimental demonstration of the beneficial effects of increased cognitively stimulating activity on fluid intelligence in healthy older
adults. Neuropsychology, Development, and Cognition, Section B: Aging, Neuropsychology, and Cognition, 15, 184 –207. doi:10.1080/
13825580701322163
Uttal, D. H., Meadow, N. G., Tipton, E., Hand, L. L., Alden, A. R., Warren,
C., & Newcombe, N. S. (2013). The malleability of spatial skills: A
meta-analysis of training studies. Psychological Bulletin, 139, 352– 402.
doi:10.1037/a0028446
Verhaeghen, P., & Marcoen, A. (1996). On the mechanisms of plasticity in
young and older adults after instruction in the method of loci: Evidence
for an amplification model. Psychology and Aging, 11, 164 –178. doi:
10.1037/0882-7974.11.1.164
Verhaeghen, P., Marcoen, A., & Goossens, L. (1992). Improving memory
performance in the aged through mnemonic training: A meta-analytic
study. Psychology and Aging, 7, 242–251. doi:10.1037/0882-7974.7.2
.242
von Bastian, C. C., & Oberauer, K. (2013). Distinct transfer effects of
training different facets of working memory capacity. Journal of Memory and Language, 69(1). doi:10.1016/j.jml.2013.02.002
Wechsler, D. (1997a). Wechsler Adult Intelligence Scale (3rd ed.). San
Antonio, TX: NCS Pearson.
Wechsler, D. (1997b). Wechsler Memory Scale (3rd ed.). San Antonio, TX:
The Psychological Corporation.
Weinfurt, K. P. (2002). Repeated measures analysis: ANOVA, MANOVA,
and HLM. In L. G. Grimm & P. R. Yarnold (Eds.), Reading and
understanding more multivariate statistics (pp. 317–361). Washington,
DC: American Psychological Association.

11

Wiley, J., & Jarosz, A. F. (2012). How working memory capacity affects
problem solving. In B. H. Ross (Ed.), Psychology of learning and
motivation (Vol. 56, pp. 185–227). Burlington, VT: Academic Press.
Wiley, J., Jarosz, A. F., Cushen, P. J., & Colflesh, G. J. H. (2011). New rule
use drives the relation between working memory capacity and Raven’s
Advanced Progressive Matrices. Journal of Experimental Psychology
Learning, Memory, and Cognition, 37, 256 –263. doi:10.1037/a0021613
Ybarra, O., Burnstein, E., Winkielman, P., Keller, M. C., Manis, M., Chan,
E., & Rodriguez, J. (2008). Mental exercising through simple socializing: Social interaction promotes general cognitive functioning. Personality and Social Psychology Bulletin, 34, 248 –259. doi:10.1177/
0146167207310454
Yesavage, J. A., & Sheikh, J. I. (2008). Geriatric Depression Scale (GDS):
Recent evidence and development of a shorter version. Clinical Gerontologist, 5, 165–173. doi:10.1300/J018v05n01_09
Zelinski, E. M. (2009). Far transfer in cognitive training of older adults.
Restorative Neurology and Neuroscience, 27, 455– 471. doi:10.3233/
RNN-2009-0495
Zhao, X., Wang, Y., Liu, D., & Zhou, R. (2011). Effect of updating training
on fluid intelligence in children. Chinese Science Bulletin, 56, 2202–
2205. doi:10.1007/s11434-011-4553-5
Zinke, K., Zeintl, M., Eschen, A., Herzog, C., & Kliegel, M. (2012).
Potentials and limits of plasticity induced by working memory training
in old-old age. Gerontology, 58, 79 – 87. doi:10.1159/000324240

Received June 28, 2012
Revision received June 13, 2013
Accepted September 25, 2013 䡲