Memory 2

Cognition-based interventions for healthy older people and
people with mild cognitive impairment (Review)
Martin M, Clare L, Altgassen AM, Cameron MH, Zehnder F
This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library
2011, Issue 1
http://www.thecochranelibrary.com
Cognition-based interventions for healthy older people and people with mild cognitive impairment (Review)
Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
TABLE OF CONTENTS
HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Analysis 1.1. Comparison 1 healthy older adults: treatment vs no contact, Outcome 1 face-name immediate recall. . 35
Analysis 1.2. Comparison 1 healthy older adults: treatment vs no contact, Outcome 2 face-name delayed recall. . . 36
Analysis 1.3. Comparison 1 healthy older adults: treatment vs no contact, Outcome 3 visuo-spatial memory. . . . 36
Analysis 1.4. Comparison 1 healthy older adults: treatment vs no contact, Outcome 4 short-term memory. . . . . 37
Analysis 1.5. Comparison 1 healthy older adults: treatment vs no contact, Outcome 5 paired associates. . . . . . 37
Analysis 1.6. Comparison 1 healthy older adults: treatment vs no contact, Outcome 6 immediate recall. . . . . . 38
Analysis 1.7. Comparison 1 healthy older adults: treatment vs no contact, Outcome 7 delayed recall. . . . . . . 39
Analysis 2.1. Comparison 2 healthy older adults: treatment versus active control, Outcome 1 face-name immediate recall. 40
Analysis 2.2. Comparison 2 healthy older adults: treatment versus active control, Outcome 2 face-name delayed recall. 40
Analysis 2.3. Comparison 2 healthy older adults: treatment versus active control, Outcome 3 visuo-spatial. . . . . 41
Analysis 2.4. Comparison 2 healthy older adults: treatment versus active control, Outcome 4 short-term memory. . 41
Analysis 2.5. Comparison 2 healthy older adults: treatment versus active control, Outcome 5 paired associates. . . 42
Analysis 2.6. Comparison 2 healthy older adults: treatment versus active control, Outcome 6 immediate recall. . . 43
Analysis 2.7. Comparison 2 healthy older adults: treatment versus active control, Outcome 7 delayed recall. . . . 44
Analysis 3.1. Comparison 3 MCI: treatment vs no contact, Outcome 1 immediate recall. . . . . . . . . . . 44
Analysis 3.2. Comparison 3 MCI: treatment vs no contact, Outcome 2 delayed recall. . . . . . . . . . . . 45
Analysis 3.3. Comparison 3 MCI: treatment vs no contact, Outcome 3 executive function. . . . . . . . . . 45
Analysis 4.1. Comparison 4 MCI: treatment vs alternative treatment, Outcome 1 immediate recall. . . . . . . 46
Analysis 4.2. Comparison 4 MCI: treatment vs alternative treatment, Outcome 2 delayed recall. . . . . . . . 46
HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
DIFFERENCES BETWEEN PROTOCOL AND REVIEW . . . . . . . . . . . . . . . . . . . . . 48
Cognition-based interventions for healthy older people and people with mild cognitive impairment (Review) i
[Intervention Review]
Cognition-based interventions for healthy older people and

people with mild cognitive impairment
Mike Martin1 , Linda Clare2 , Anne Mareike Altgassen3 , Michelle H Cameron4 , Franzisca Zehnder1
1 Psychologisches Institut, Universität Zürich, Lehrstuhl Gerontopsychologie, Zürich, Switzerland. 2 School of Psychology, University
of Wales Bangor, Bangor, UK. 3 University of Dresden, Dresden, Germany. 4 Department of Neurology, Oregon Health and Science
University, Portland, Oregon, USA
Contact address: Mike Martin, Psychologisches Institut, Universität Zürich, Lehrstuhl Gerontopsychologie, Binzmühlestrasse 14/24,
Zürich, CH-8050, Switzerland. m.martin@psychologie.uzh.ch.
Editorial group: Cochrane Dementia and Cognitive Improvement Group.

Publication status and date: New, published in Issue 1, 2011.
Review content assessed as up-to-date: 2 January 2009.
Citation: Martin M, Clare L, Altgassen AM, Cameron MH, Zehnder F. Cognition-based interventions for healthy older people
and people with mild cognitive impairment. Cochrane Database of Systematic Reviews 2011, Issue 1. Art. No.: CD006220. DOI:
10.1002/14651858.CD006220.pub2.
ABSTRACT
Background
Evidence from some, but not all non-randomised studies suggest the possibility that cognitive training may influence cognitive
functioning in older people. Due to the differences among cognitive training interventions reported in the literature, giving a general
overview of the current literature remains difficult.
Objectives
To systematically review the literature and summarize the effect of cognitive training interventions on various domains of cognitive
function (ie memory, executive function, attention and speed) in healthy older people and in people with mild cognitive impairment.
Search strategy
The CDCIG Specialized Register was searched on 30 September 2007 for all years up to December 2005. The Cochrane Library,
MEDLINE, EMBASE, PsycINFO and CINAHL were searched separately on 30 September 2007 to find trials with healthy people.
These results were supplemented by searches from January 1970 to September 2007 in PsychInfo/Psyndex, ISI Web of Knowledge and
PubMed.
Selection criteria
RCTs of interventions evaluating the effectiveness of cognitive training for healthy older people and people with mild cognitive
impairment from 1970 to 2007 that met inclusion criteria were selected.
Data collection and analysis
Authors independently extracted data and assessed trial quality. Meta-analysis was performed when appropriate.
Cognition-based interventions for healthy older people and people with mild cognitive impairment (Review) 1
Main results
Only data on memory training could be pooled for analysis. Within this domain, training interventions were grouped according to
several outcome variables. Results showed that for healthy older adults, immediate and delayed verbal recall improved significantly
through training compared to a no-treatment control condition. We did not find any specific memory training effects though as the
improvements observed did not exceed the improvement in the active control condition. For individuals with mild cognitive impairment,
our analyses demonstrate the same pattern. Thus, there is currently little evidence on the effectiveness and specificity of memory
interventions for healthy older adults and individuals with mild cognitive impairment.
Authors’ conclusions
There is evidence that cognitive interventions do lead to performance gains but none of the effects observed could be attributable
specifically to cognitive training, as the improvements observed did not exceed the improvement in active control conditions. This does
not mean that longer, more intense or different interventions might not be effective, but that those which have been reported thus far
have only limited effect. We therefore suggest more standardized study protocols in order to maximize comparability of studies and to
maximize the possibility of data pooling - also in other cognitive domains than memory.
PLAIN LANGUAGE SUMMARY
Effects of memory training in healthy older adults and older adults with mild cognitive impairment
There is an increasing interest in information on the effectiveness of cognitive training interventions to improve memory in normal
and mildly cognitively impaired older adults (60 years and older). We analyzed all cognitive interventions between 1970 and 2007 to
determine their effectiveness. The results suggest that cognitive interventions do lead to performance improvements and that the size of
the effects differs for different kinds of memory skills in healthy older adults and people with mild cognitive impairment. In particular,
immediate and delayed verbal recall improved significantly through training compared to a no-treatment control condition but the
improvements observed did not exceed the improvement in the active control conditions.
BACKGROUND
of studies have indicated that cognitively-stimulating activity may
As our societies age, and at the same time become more techno- help to protect against cognitive decline in later life (Wilson 2002).
logically complex, there is increasing interest in understanding the Building on these observations, researchers have attempted to en-
effects of ageing on cognitive function. We draw on the range of hance or maintain cognitive functioning in older people by means
abilities in areas such as attention, perception, memory, and lan- of systematic cognition-based interventions such as memory train-
guage for many activities in our daily lives. Most people, although ing. It is important to establish to what extent cognitive perfor-
not all, experience a cognitive decline in old age. There is, how- mance can be improved through systematic training across adult-
ever, also evidence for potential gains in performance, in particu- hood and old age, and for how long any gains are maintained. It
lar in domains where performance is supported by greater expe- is also important to establish what factors influence the extent of
rience. This can be demonstrated in a number of areas of exper- any gains for a given individual, and to determine how different
tise, ranging from vocabulary to job-specific skills and knowledge. features of the training, such as intensity, frequency, duration, or
These findings show that there is potential for cognitive plasticity focus, impact on the size of the gains (Hoyer 2006; Nyberg 2005;
(change and adaptation) in later life (Hoyer 2006; Kliegl 1989; Willis 2001).
Verhaeghen 1992), whereby performance can be enhanced under
optimal conditions (Singer 2003). The possible extent and limits
For the majority of older people the extent of any cognitive decline
of cognitive plasticity in later life remain to be determined.
is relatively small, but some individuals develop more extensive
Understanding more about the processes underlying these changes difficulties and are at greater risk of developing a form of dementia.
in cognitive performance may offer various avenues for support- Various terms and definitions have been applied to this group; cur-
ing cognitive functioning in later life. Findings from a number rently, they are likely to be described as experiencing ’mild cogni-
tive impairment’. Individuals with MCI display cognitive changes Numerous studies report the effects of cognition-focused inter-
that are not severe enough to fulfil diagnostic criteria for dementia, ventions with older people. There is some evidence for cognitive
but are greater than those typically observed in their age group plasticity in later life as well as a possible protective effect of en-
(Larrieu 2002; Petersen 2001). Earlier definitions emphasize the gaging in cognitively-stimulating activity. This suggests there may
differentiation from optimal ageing (e.g. “Benign Senescent For- be potential to improve cognitive functioning in later life through
getfulness”; Kral 1962; “Age-Associated Memory Impairment”; cognitive training interventions, and this in turn might help to
AAMI; Crook 1986), or the identification of preclinical demen- support continued independence and maximise quality of life for
tia patients (e.g. “Malignant Senescent Forgetfulness”; Kral 1962; otherwise healthy older people. For older people who are already
“Cognitive Impairment, No Dementia”; CIND; Graham 1997). experiencing mild cognitive impairment, and who are at increased
The term MCI as defined by the American Psychiatric Association risk of developing dementia, cognition-focused interventions may
(APA 1987) is a condition involving impaired short- and long- help to improve or maintain the level of cognitive performance
term memory, but no functional impairment. MCI is assumed to and thereby delay or prevent further decline (Hoyer 2006; Wilson
be a precursor of dementia, i.e. a transitional state between normal 2002).
cognitive decline in old age and dementia. Due to the variability The most frequently reported form of cognition-focused interven-
in definitions, studies investigating prevalence and incidence of tion is cognitive training. Cognitive training involves individual
MCI come to different conclusions (Kratz 2002). Prevalence rates or group sessions with practice on tasks targeting aspects of cog-
vary between 5% and 25% (Kumar 2005; Manly 2005; Purser nitive functioning such as memory, attention and language. The
2005), incidence rates between 0.5 and 8% (Busse 2003; Larrieu precise parameters of cognitive training interventions reported in
2002; Jungwirth 2005). the literature vary considerably, and as a result it has been difficult
to draw firm conclusions about efficacy. This review aims to gain a
Older people with mild cognitive impairment constitute a par-
clearer picture of the effectiveness of cognitive training, in order to
ticularly vulnerable, at-risk group. Cognition-based interventions
provide guidance on when to apply which training to whom and
may offer the possibility of maintaining or improving cognitive
how often in order to achieve the greatest benefits. Effectiveness
function, and perhaps prevent or delay progression to dementia
can be considered in terms of improvements on test scores in the ar-
(Hultsch 1999; Schooler 2001; Stern 2002; Unverzagt 2007). It
eas of cognitive functioning targeted in the training, maintenance
is also important to determine whether the possible benefits differ
of improvements over time, transfer of training effects to other
from those seen in healthy older people, and whether the same or
kinds of cognitive tasks, and generalisation of effects to everyday
different forms of intervention are most suitable (Nyberg 2005).
functioning. It is also important to consider what factors may be
responsible for any benefits resulting from cognitive training, and
whether the same, or different approaches are needed for healthy
Intervention older people and older people with mild cognitive impairment.
This review will assess the effectiveness of cognitive training. Cog- Cognition-based interventions such as memory training have fo-
nitive training is defined as an intervention providing structured cused on examining the potential for improvement of cognitive
practice on tasks relevant to aspects of cognitive functioning, such functioning in normal ageing and on determining the limits of cog-
as memory, attention, language or executive function. Standard- nitive plasticity in old age (Hoyer 2006; Kliegl 1989; Verhaeghen
ized tasks are used (Clare 2003) but level of difficulty may be 1992). Cognitive plasticity refers to cognitive changes and adapta-
graded to allow for individual variations in ability. The selected tions, and especially to the possible performance of people under
tasks vary in degree of specificity, with some interventions focus- optimal conditions (Singer 2003).
ing on very specific abilities and strategies, and others taking a Current practice in cognition-based interventions includes group
more multimodal and holistic approach. Cognitive training may training targeting memory, attention and language. Cognitive
be offered in various forms, including individual or group sessions, training interventions may address individuals or groups. They
and tasks may be presented in various modalities, including pen- differ with regards to trained abilities (e.g. memory, attention,
cil-and-paper or computerised versions. There is wide variation speed of information processing), specificity of training (e.g. train-
in frequency and duration of training sessions. This intervention ing of text recall vs. multimodal and holistic approaches training
approach is intended to address cognitive function and/or cogni- a combination of abilities), strategies practiced in the training ses-
tive impairment directly and to produce improvements in perfor- sions (e.g. method of loci, imagery training), duration of training
mance on standardised measures of the relevant domains. Effects sessions and overall training period, frequency of training sessions,
on performance of specific tasks trained in the intervention may group size and participant characteristics (e.g. education, person-
also be considered. ality, preferred learning style etc.).
Key questions in cognition-based interventions are the range of
potential improvements in essential areas of cognitive function-
Rationale ing, maintenance of improvements, and transfer of training effects
to everyday functioning. Understanding the factors responsible • Either healthy older people with no diagnosis or older
for improvements provides the possibility of making cognitive in- people who meet criteria for mild cognitive impairment;
terventions more cost-effective. In addition, differences in effects Peterson criteria for MCI were used; depressive symptoms were
and optimal training methods between normal older adults and excluded by administration of GDS or Profile of Mood States.
adults with mild cognitive impairments can be examined. Findings • Normally educated
might suggest that preventive interventions at earlier ages might • Participants with a diagnosis of dementia were excluded
be promising, but need to be examined for long-term effects. • Profiles of general cognitive ability and cognitive
Due to the differences among cognitive training interventions re- functioning in relevant domains, as indicated by performance on
ported in the literature, giving a general overview of the current standardised measures, must be documented to allow an
literature remains difficult. Moreover, conclusions of studies are evaluation of participants’ cognitive status and, specifically,
based on different designs and outcomes, such as pre-post com- whether they fit the definition of mild cognitive impairment. In
parisons, randomized control groups or comparisons with alter- order not to exclude studies that may be relevant for this review,
native trainings (active controls). The present review aims to gain none of the specific definitions of mild cognitive impairment are
a clearer picture of the effectiveness of cognitive training, in order particularly included or excluded, but information on
to provide guidance on when to apply which training to whom participants’ cognitive ability is required for classification of
and how often in order to optimize efficacy. individual cognitive status. We need information on participants’
memory and general cognitive ability in comparison to norms to
be able to classify participants on individual cognitive status.
OBJECTIVES
Types of interventions
The purpose of the present review is to evaluate the effectiveness Studies were considered for this review if they describe cognitive
of cognitive training in healthy older adults and older adults with training interventions targeting specific domains of cognitive func-
mild cognitive impairment. Therefore, studies examining cogni- tioning such as memory, attention, or speed. No contact control /
tive training with the above mentioned target groups are analysed no treatment will be defined as no training, and active control con-
with regards to training effectiveness and (if possible) sustainabil- ditions will comprise non-cognitive activities, unspecific cognitive
ity. This review will help practitioners to choose suitable training stimulation, such as art discussion (Best, Hamlett & Davis, 1992)
methods and may inform future research. and alternative or active control training (e.g., attention training;
Scogin & Prohaska, 1992). Intervention settings were individual
or group settings.
METHODS When several control groups were compared to the treatment
group, e.g., no treatment and multiple alternative or active con-
trols, we considered only one comparison group respectively in the
Criteria for considering studies for this review two possible comparison conditions (no contact and altervative
treatment).
Duration of intervention was up to one year, with at least a baseline
Types of studies and a post-intervention assessment reported.
To date, most studies investigating the effectiveness of cognitive

training have used pre-post designs or relied on comparisons with Types of outcome measures
alternative approaches, active control conditions or waiting list
• For the MCI group, rates of conversion to dementia and, if
control conditions.
applicable, rates of institutionalisation were considered, but none
This review focuses on randomized control trials (RCTs), for which
of the included studies provided this information;
adequate information was provided. The minimum number of
• Incidence and severity of adverse effects were considered,
measurements and assessments is two. The studies included must
but none of the included studies provided this information.
have been published, written in English or German, and presented
in a journal article, to avoid the situation where the same or highly The following cognitive variables were considered as outcome
related data was reported in both journals and book chapters. measures: any measures of cognitive functioning, improvement,
sustainability and transfer of training effects. The most impor-
tant cognitive indicators were immediate and delayed recall of
Types of participants face-name associations, visuo-spatial memory, short term mem-
• Participants (both male and female) aged 60 years or older ory, paired associates, and immediate and delayed recall (of words,
• Any setting (group and individual) paragraphs, stories). Outcome variables such as well-being, quality
of life and everyday functioning were considered as non-cognitive • CINAHL (1982 to 2006/06);
outcomes and were therefore not included. • SIGLE (Grey Literature in Europe) (1980 to 2005/03);
Desirable outcome information from the studies relates to im- • LILACS: Latin American and Caribbean Health Science
provement of participants’ performance on the trained variables. Literature (http://bases.bireme.br/cgi-bin/wxislind.exe/iah/
Improved cognitive functioning might delay onset of pathological online/?IsisScript=iah/iah.xis&base=LILACS&lang=i&form=F)
cognitive decline in old age or lessen the burden that impairment (last searched 29 August 2006).
places on participants and significant others. Preferably, continu-
ous scales should be used to be able to assess the full range. Studies
were only included in the review if they recorded participants’ per- Conference roceedings
formance at least at two time points (before and after the training). • ISTP (http://portal.isiknowledge.com/portal.cgi) (Index to
Scientific and Technical Proceedings) (to 29 August 2006);
• INSIDE (BL database of Conference Proceedings and
Search methods for identification of studies Journals) (to June 2000)
See Cochrane Dementia and Cognitive Improvement Group

methods used in reviews. Theses
The Specialized Register of the Cochrane Dementia and Cogni- • Index to Theses (formerly ASLIB) (http://www.theses.com/
tive Improvement Group (CDCIG) was searched on 30 Septem- ) (UK and Ireland theses) (1716 to 11 August 2006);
ber 2007 for all years up to December 2005. This register contains • Australian Digital Theses Program (http://adt.caul.edu.au/
records from the major healthcare databases The Cochrane Library, ): (last update 24 March 2006);
MEDLINE, EMBASE, PsycINFO, CINAHL and LILACS, and • Canadian Theses and Dissertations (http://
many ongoing trial databases and other grey literature sources. www.collectionscanada.ca/thesescanada/index-e.html): 1989 to
The following search terms were used: ’cognitive stimulation’ OR 28 August 2006);
’cognitive rehabilitation’ OR ’cognitive training’ OR ’cognitive • DATAD - Database of African Theses and Dissertations
retraining’ OR ‘cognitive re-training’ OR ’cognitive support’ OR (http://www.aau.org/datad/backgrd.htm);
’memory function’ OR ’memory rehabilitation’ OR ’memory ther- • Dissertation Abstract Online (USA) (http://
apy’ OR ’memory aid*’ OR ’memory group*’ OR ’memory train- wwwlib.umi.com/dissertations/gateway) (1861 to 28 August
ing’ OR ’memory retraining’ OR ’memory support’ OR ’memory 2006).
stimulation’ OR ’memory strategy’ OR ’memory management’.
The Cochrane Library, MEDLINE, EMBASE, PsycINFO and
CINAHL were searched separately on 30 September 2007 to find Ongoing trials
trials with healthy people. The following search terms were used:
’cognitive stimulation’ OR ’cognitive rehabilitation’ OR ’cogni-
tive training’ OR ’cognitive retraining’ OR ‘cognitive re-training’ UK
OR ’cognitive support’ OR ’memory function’ OR ’memory re- • National Research Register (http://www.update-
habilitation’ OR ’memory therapy’ OR ’memory aid*’ OR ’mem- software.com/projects/nrr/) (last searched issue 3/2006);
ory group*’ OR ’memory training’ OR ’memory retraining’ OR • ReFeR (http://www.refer.nhs.uk/ViewWebPage.asp?Page=
’memory support’ OR ’memory stimulation’ OR ’memory strat- Home) (last searched 30 August 2006);
egy’ OR ’memory management’. These search terms were used • Current Controlled trials: Meta Register of Controlled trials
in combination with Phases 1 to 3 of the Highly sensitive search (mRCT) (http://www.controlled-trials.com/) (last searched 30
strategies for identifying reports of randomized controlled trials in August 2006)
MEDLINE (APPENDIX 5b, Cochrane Handbook, 2006), and • ISRCTN Register - trials registered with a unique identifier
all terms were searched as Title, abstract, keyword, Publication • Action medical research
type. • Kings College London
On 30 September 2007, the Register consisted of records from the • Laxdale Ltd
following databases: • Medical Research Council (UK)
• NHS Trusts Clinical Trials Register
• National Health Service Research and Development Health
Healthcare databases Technology Assessment Programme (HTA)
• CENTRAL: (The Cochrane Library 2006, Issue 1); • National Health Service Research and Development
• MEDLINE (1966 to 2006/07, week 5); Programme ’Time-Limited’ National Programmes
• EMBASE (1980 to 2006/07); • National Health Service Research and Development
• PsycINFO (1887 to 2006/08, week 1); Regional Programmes
• The Wellcome Trust The search terms used were: ’memory training’, ’mnemonic train-
• Stroke Trials Registry (http://www.strokecenter.org/trials/ ing’, ’cognitive training’, ’cognitive rehabilitation’, ’cognitive in-
index.aspx) (last searched 31 August 2006). tervention’, ’cognitive exercise’ in combination with ’elderly’, ’old
adults’, old age’, ‘MCI’, ‘mild cognitive impairment’, ’memory
complainers’, ’AACD’, ’dementia’, ’dementia treatment’, and ’de-
Netherlands mentia therapy’. After searches were completed among the major
databases, reference lists from acquired studies and recent meta-
• Nederlands Trial Register (http://www.trialregister.nl/
analyses were examined to find additional RCTs.
trialreg/index.asp) (last searched 31 August 2006).
USA/International
Data collection and analysis
• ClinicalTrials.gov (http://www.ClinicalTrials.gov) (last Searches were conducted as detailed above to identify all relevant
searched 31 August 2006) (contains all records from http:// published studies, and hard copies of articles were obtained. RCTs
clinicalstudies.info.nih.gov/); were identified and four reviewers (MM, MA, FZ and LC) worked
• IPFMA Clinical trials Register: www.ifpma.org/ independently to determine which studies meet the criteria for
clinicaltrials.html. The Ongoing Trials database within this inclusion before reaching a final consensus on which studies to
Register searches http://www.controlled-trials.com/isrctn, http:// include.
www.ClinicalTrials.gov and http://www.centerwatch.com/. The
ISRCTN register and Clinicaltrials.gov are searched separately. Quality assessment
Centerwatch is very difficult to search for our purposes and no
update searches have been done since 2003. The reviewers assessed the methodological quality of randomiza-
• The IFPMA Trial Results databases searches a wide variety tion in each trial using one of the approaches described in the
of sources among which are: Cochrane Reviewers’ Handbook (Higgins 2008):
• http://www.astrazenecaclinicaltrials.com (seroquel, statins) In category A (adequate), the report describes allocation of treat-
• http://www.centerwatch.com ment by: (i) some form of centralized randomized scheme, such
• http://www.clinicalstudyresults.org as having to provide details of an enrolled participant to an of-
• http://clinicaltrials.gov fice by telephone to receive the treatment group allocation; (ii)
• http://www.controlled-trials.com some form of randomization scheme controlled by a pharmacy;
• http://ctr.gsk.co.uk (iii) numbered or coded containers, as in a pharmaceutical trial in
• http://www.lillytrials.com (zyprexa) which capsules from identical-looking numbered bottles are ad-
• http://www.roche-trials.com (anti-abeta antibody) ministrated sequentially to enrolled participants; (iv) an on-site or
• http://www.organon.com coded computer system, provided that the allocations were in a
• http://www.novartisclinicaltrials.com (rivastigmine) locked, unreadable file that could be accessed only after inputting
• http://www.bayerhealthcare.com the characteristics of an enrolled participants; or (v) if assignment
• http://trials.boehringer-ingelheim.com envelopes were used, the report should at least specify that they
• http://www.cmrinteract.com were sequentially numbered, sealed, and opaque; (vi) other com-
• http://www.esteve.es binations of described elements of the process that provide assur-
• http://www.clinicaltrials.jp ance of adequate concealment.
Category B (intermediate) is where the report describes allocation
This part of the IPFMA database is searched and was last updated of treatment by: (i) use of a ”list” of ”table” to allocate assignments;
on 4 September 2006; (ii) use of ”envelopes” or ”sealed envelopes”; (iii) stating the study
• Lundbeck Clinical Trial Registry (http:// as ”randomized” without further detail.
www.lundbecktrials.com) (last searched 15 August 2006); Category C (inadequate) is where the report describes allocation
• Forest Clinical trial Registry (http:// of treatment by: (i) alternation; (ii) reference to case record num-
www.forestclinicaltrials.com/) (last searched 15 August 2006). bers, dates of birth, day of week, or any such approach; (iii) any
allocation procedure that is transparent before assignment, such
The search strategies used to identify relevant records in MED- as an open list of random numbers or assignments. Empirical re-
LINE, EMBASE, PsycINFO, CINAHL and LILACS can be search has shown that lack of adequate allocation concealment
found in the Group’s module on The Cochrane Library. is associated with bias. Trials with unclear concealment measures
These results were supplemented by searches from January 1970 have been shown liable to yield more pronounced estimates of
to September 2007 in PsyhInfo/Psyindex, ISI Web of Knowledge treatment effects than trials that have adequate measure to con-
and PubMed. ceal allocation schedules, but the effect is less pronounced than
inadequately concealed trials (Chalmers 1983; Schulz 1995). Tri- difference divided by the pooled standard deviation when they
als were considered if they conformed to categories A or B, but used different rating scales or tests.
those falling in category C were excluded. Other aspects of trial The duration of the trials varied considerably. Some training inter-
quality were not assessed by a scoring system but details of blind- ventions covered equally long time spans, but differed in intensity
ing, appropriateness of methods and the number of patients lost or vice versa. If one assumes that the intensity and frequency is
to follow-up were noted. the most important determinant of the occurrence of a training
effect, then the difference in time span might be neglected as long
as time spans do not exceed several months (as provided for in our
Data extraction inclusion criteria). Thus, we decided to combine all trials into the
Data from the RCTs selected for inclusion was extracted. The respective meta-analyses to maximize the information extracted
summary statistics required for each trial and each outcome for from the database. Once more training data are available, it might
continuous data are the mean change from baseline, the standard be appropriate to divide the studies into smaller time periods and
error of the mean change, and the number of patients for each to conduct a separate meta-analysis for studies of different dura-
treatment group at each assessment. Where changes from baseline tions. Some trials might contribute data to more than one time
were not reported, the mean, standard deviation and the number period if multiple assessments have been made.
of people in each treatment group at each time point was extracted We selected one variable from each study to represent the outcome
if available. measure and when several control groups were compared to the
The baseline assessment is defined as the latest available assessment treatment group (i.e. several active control groups), we selected
prior to randomization, but no longer than two months before. only one group for comparison. This has the advantage of not giv-
For each outcome measure, data of those who completed the trial ing too much weight to one study but does minimize information
was sought and indicated as such. Wherever possible, the data extraction from the database.
were sought irrespective of compliance, whether or not the person For binary outcomes, such as improvement or no improvement,
was subsequently deemed ineligible, or otherwise excluded from the odds ratio was used to measure treatment effect. A weighted
treatment or follow-up. estimate of the typical treatment effect across trials was calculated.
Overall estimates of the treatment difference are presented. In all
cases the overall estimate from a fixed-effects model is presented
Data analysis and a test for heterogeneity using a standard chi-square statistic
was performed. If, however, there is evidence of heterogeneity of
We pooled studies with sufficient data, judged to be clinically
the treatment effect between trials then either only homogeneous
homogeneous, using RevMan 5.0 software.
results will be pooled, or a random-effects model used (in which
We intended to include studies addressing or cognitive domains
case the confidence intervals would be broader than those of a
other than memory, but no more than one study in any given
fixed-effects model). When studies were statistically heterogeneous
domain was identified that met our inclusion criteria.
(I2 test value > 50%), a random-effect model was used; otherwise
The outcomes measured may arise from ordinal rating scales.
a fixed-effect model was used.
Where the rating scales used in the trials have a reasonably large
number of categories (more than 10) the data were treated as con-
tinuous outcomes arising from a normal distribution.
Summary statistics (n, mean and standard deviation) were used
RESULTS
for each rating scale at each assessment time for each treatment
group in each trial for change from baseline. For cross-over trials
only the data from the first treatment period was used.
When change from baseline results were not reported, the required
Description of studies
summary statistics were calculated from the baseline and assess- See: Characteristics of included studies; Characteristics of excluded
ment time treatment group means and standard deviations. In studies.
this case a zero correlation between the measurements at baseline From the initial set of references identified by the systematic
and assessment time was assumed. This method overestimates the searches, a set of thirty-six studies met the inclusion criteria. Two
standard deviation of the change from baseline, but this conserva- studies were ranked as grade A and thirty-four as grade B. Thirty-
tive approach is considered to be preferable in a meta-analysis. three of the included studies involved healthy older people and
Meta-analysis requires the combination of data from trials that may three of them investigated people with mild cognitive impairment.
not use the same rating scale to assess an outcome. The measure Sample size, patients’ age, type of cognitive training, type of con-
of the treatment difference for any outcome is the weighted mean trol condition, duration and modality of training, outcome mea-
difference when the pooled trials use the same rating scale or test, sures, and effect sizes were evaluated and are presented in the tables
and the standardized mean difference, which is the absolute mean on characteristics of included and excluded studies.
Overall, 767 healthy older adults and 34 participants with mild Results of the search
cognitive impairment, 442 no contact controls and 986 controls We identified 36 randomized controlled trials including a total of
with alternative treatments (active controls) were included in the 2229 participants with an estimated mean age of 69.90 years (SD
analysis. 3.53) (mean age was estimated from midpoint of the age range for
The included studies varied in many aspects. They varied consid- those studies in which mean age was not reported).
erably in terms of number of training sessions and overall duration Interventions were grouped into cognitive domains (such as mem-
of the intervention: the time devoted to training sessions varied ory, executive function, attention and speed) and then pooled to
between 6 and 135 hours, and the overall period of the cognition- create ability subgroups within the domains that were as homoge-
based interventions between one day and one year. Less divergent, neous as possible. Studies providing data on training in speed of
but still variable and not always indicated, were pre- to post-test processing, attention and executive functioning were excluded due
intervals and training to post-test intervals. The post-treatment- to lack of correspondence to inclusion criteria. Therefore, these
assessments took mainly place immediately after training (imme- domains could not be assessed and only data on memory training
diately after training completion or within one week after com- could be pooled for analysis.
pletion). The duration of the trainings varied between a few hours Within the memory domain, training interventions were grouped
and a year. Most of the interventions were conducted in a group according to the following ability subgroups: face-name immedi-
setting with a trainer or tutor, and a minority were self-instruc- ate and delayed recall, visuo-spatial memory, short term memory,
tional or conducted on an individual basis. Cognition-based train- paired associate learning, immediate recall and delayed recall. Data
ing intervention groups focused primarily on mnemotechniques on prospective memory could not be pooled. Therefore, for stud-
and multifactorial training which combined various methods. Ac- ies with healthy older people, only data on immediate and delayed
tive alternative treatments included group discussions as well as (face-name) recall, visuo-spatial memory, short-term memory, and
physical training and drug treatment alone or in combination with paired associate learning could be pooled. Since only three studies
strategy training. No contact control groups had no training at all. included people with mild cognitive impairment, data pooling for
this group was only possible for one of the seven outcome mea-
sures in the memory domain (immediate recall) (Table 1).
Cognitive Domain Ability subgroups Availability of pooled data for
healthy older people People with MCI
Memory face-name immediate recall yes not enough data
face-name delayed recall yes not enough data
visuo-spatial memory (Benton, vi- yes not enough data

sual reproductions)
short term memory (digit span) yes not enough data
paired associates yes not enough data
immediate recall (of words, para- yes yes

graphs, stories)
delayed recall (of words, para- yes not enough data

graphs, stories)
prospective memory not enough data not enough data
Executive Functions semantic verbal fluency not enough data not enough data
(Continued)
phonematic verbal fluency not enough data not enough data
abstraction not enough data not enough data
inductive reasoning not enough data not enough data
Attention - not enough data not enough data
Speed - not enough data not enough data
Table 1: Overview of evaluated cognitive domains and ability

subgroups. over a 37-year period examining the effects of cognitive training
using a randomized control design. Either it is difficult to publish
Included studies the replication of an existing finding (publication bias) or the goal
of most training studies is to demonstrate that individuals improve
Thirty-six studies were selected for inclusion in the current meta- after training and determining the cause for the improvement is of
analysis. These studies a) included healthy normal participants or secondary interest (as demonstrated by a relatively large number
people with mild cognitive impairment, b) included a pre- and of excluded studies). As a consequence, published studies are more
posttreatment measure of memory performance and other cogni- likely to contain significant improvements after training. Second,
tive domains (studies that investigated long-term effects through there might be a bias towards overestimating effects by using the
follow-up were not included), and c) provided sufficient statisti- treatment for multiple comparisons with no contact controls and
cal data for the computation of effect sizes. A total of 34 research one or more active control conditions. Third, outcome variables
papers was retrieved, including information on 36 studies. How- are pooled that measure the same ability but are not totally iden-
ever, pooling of studies was only possible where at least two studies tical. Fourth, only two of the included studies reported an ade-
covered the same intervention domain. Thus, 24 studies could be quate allocation concealment or double-blind design, whereas the
pooled with regard to outcome measures. other studies were described as “randomized” without further de-
tail. Considering these types of biases and the limited evidence for
Excluded studies positive effects after combination of data in a meta-analysis, even
these improvements might be overestimating the actual chances
Over 120 studies were excluded because they focused on patients
of improvement through cognitive training interventions.
with cognitive impairments or a diagnosis of dementia. Thirty-
nine studies were excluded because they were not intervention
studies, were reviews, were not journal articles, or were written
in neither English nor German. Reasons for excluding the rest of Effects of interventions
the studies were as follows (see table “Characteristics of exluded
studies”): For calculations we used RevMan 5 (see data analysis).
a) non-randomised study design (45) We identified 36 randomized controlled trials, and 24 were finally
b) no pre-/post design (3) pooled including a total of 2229 participants. Interventions were
c) age range (7) grouped into cognitive domains and only data on memory training
d) missing data (17) could be pooled. Therefore within the memory domain, training
interventions were grouped according to outcome variables: face-
name immediate and delayed recall, visuo-spatial memory, short
Risk of bias in included studies term memory, paired associates, immediate recall and delayed re-
As is typical for meta-analyses on the effectiveness of interven- call.
tions, there are several sources of bias in published training stud- For healthy older adults, immediate and delayed verbal recall im-
ies. First, there is a bias toward publishing studies or publishing proved significantly (p<0.05) through training compared to a no-
results from test instruments demonstrating significant gains after treatment control condition.
training. This may have led to the relatively small set of 36 studies For individuals with mild cognitive impairment, our analyses
demonstrate that significant training gains were obtained for treat- were significantly better for treatment compared to no contact
ment compared to no contact control in immediate (p=0.04) re- control in only two of the seven cognitive domains with sufficient
call and delayed recall (p=0.05) and this improvement was also data for meta-analysis, namely immediate and delayed recall. Im-
not specific as it did not exceed the improvement observed in the provements were not specific, because they were no larger than
active control condition (no data available in the active control those seen in the active control conditions.
condition for delayed recall).
For individuals with mild cognitive impairment, the available data
were scarce. Most included studies used the Petersen criteria to as-
sess participants with MCI but differences in the exact application
of this definition may exist e.g. in terms of exclusion of depression
DISCUSSION and vascular risk factors (some use GDS scale, others the Profile
of Mood States), the use of MMSE-scores (one of the included
Before drawing firm conclusions concerning the results of this
studies used Mini Mental State Examinatin scores greater than or
analysis, a number of caveats need to be mentioned. First, con-
equal to 24, the other used scores greater than or equal to 25),
sidering the large time span covered, surprisingly few studies were
the use of a psychometric criterion, e.g. >1.5 standard deviations
identified that fulfilled the relatively flexible inclusion criteria. As
below expected performance for age, or a criterion based on a
a consequence, although our focus was on cognitive training in
clinical interview. Our analyses demonstrate significant training
general, most of the included studies included focus on memory
gains. However, the effects were significantly better for treatment
training interventions, and very few on speed improvements or
compared to no contact control in one outcome measures with
training of executive functioning. This might have to do with the
sufficient data for meta-analysis, namely immediate recall. This
fact that speed improvements or improvements in executive func-
improvement was also not specific as it did not exceed the im-
tions might require more intensive or extensive training, but there
provement from the active control condition. Thus, it seems that
are scarcely any studies on the effects of cognitive interventions
alternative interventions do just as well as cognitive interventions,
lasting longer than six months. Indeed, there were too few studies
and the training interventions cannot be regarded as effective be-
to allow us to calculate meta-analyses for these domains.
cause they do not improve on the effects of active control condi-
Second, in terms of the reasons for not fulfilling the inclusion tions.
criteria, most critical were (a) non-availability of complete infor-
mation, for example about the participant recruitment, the exact
procedure, and how temporary non-compliance was dealt with AUTHORS’ CONCLUSIONS
and (b) lack of a control condition. Judging from the results ob-
tained that would typically be biased towards reporting studies Implications for practice
with strong effects, one may speculate that effects of pre-post de-
signs are typically so small that researchers did not expect a signif- As the performance improvements observed did not exceed the
icant effect after controlling for repeated measurement of evalua- improvement in active control conditions, we did not find any
tion instruments and thus focused on reporting the improvement specific training effects for any of the abilities with sufficient data
from pre- to post-treatment. for the analysis. There is evidence that cognitive interventions tar-
geting the improvement of memory in healthy older adults and
Third, it appears that studies vary enormously, even within each people with mild cognitive impairment are effective in producing
subdomain we analysed, with respect to potentially influential fac- improvement in verbal immediate and delayed recall but that these
tors such as overall length of intervention, number of treatments, cognitive training effects are not specific, i.e., alternative interven-
group sizes, assurance of equal training procedures, combination tions (active controls) do just as well as training interventions in
of training contents within and across sessions, training and sim- mild cognitive impairment. It remains an open question at this
ilarity of trainers, or pre-existing training experience. In addition, point if the heterogeneity of the populations tested or the quality
it was not always obvious how the evaluation instruments were of the interventions may have influenced the results and we can
matched to the training contents (which would typically improve only speculate as to whether more intensive and longer training
the reported effects). Thus, when conducting the meta-analyses, may be needed to achieve effects larger than in active control con-
we decided to use the complete available information, but com- ditions.
promised on the heterogeneity of the studies included in each do-
main. Thus, future analyses of more studies may provide better Implications for research
evidence for the effects of confounding factors.
Our analyses provide surprisingly little evidence for the effective-
Overall, despite the limitations of our analyses, results show that ness and specificity of cognitive training interventions. Consider-
most interventions were effective, with significant improvements ing the sources of bias included that would typically lead to an over-
following training for the treatment group. However, the effects estimation of training effects, this argues against the effectiveness
of cognitive training interventions. However, it may also suggest such as prospective memory or goal-setting. This is reasonable be-
that future research needs to provide a more conclusive evidence cause improvements of basic abilities are prerequisite for transfer
base to make it possible to establish the effectiveness of cognitive to more complex tasks which draw on a number of these basic
interventions. First, a more standardized approach to examining abilities. The ability to adjust the use of cognitive skills to perform
the effectiveness of cognitive training is needed. Due to the hetero- more complex tasks may be better captured by focusing on indi-
geneity of procedures, durations, intensities, methods of dealing vidual learning trajectories compared to focusing on mean level
with absent training participants, use of a variety of training con- changes. Fourth, there are very few studies on the effectiveness of
tents, content combinations, and matching of evaluation instru- cognitive training interventions in individuals with mild cogni-
ments to training contents, the effects might be substantially larger tive impairment of any diagnostic kind. A consistent definition or
if more similar studies could be pooled for the meta-analyses. Sec- agreement on few core criteria of mild cognitive impairment may
ond, many training approaches include a combination of several help to gather evidence more quickly because a more widespread
elements, and trained individuals may respond quite differently to use of this definition would make this more likely a group of re-
the different elements of the training. Thus, training effects on an search interest. Variations in type and intensity of existing train-
individual level may be substantially higher than the group effects. ing interventions are needed to gain better knowledge about the
Therefore, in future research, collapsing data within individuals efficacy of cognitive interventions in mild cognitive impairment.
before aggregating on a group level might provide more appro-
priate tests of the effectiveness of cognitive interventions. Third,
there are clearly more studies reporting the effects on rather basic
abilities such as free recall compared to more complex behaviours ACKNOWLEDGEMENTS
We thank the Cochrane Dementia and Cognitive Improvement Group for their support in running the
searches and providing editorial support and advice in development of the protocol and review.
REFERENCES
References to studies included in this review Flynn 1990 {published data only}
Flynn TM, Storandt M. Supplemental group discussions in
Ball 2002 {published data only} memory training for older adults. Psychology and Aging 1990;5(2):
Ball K, Berch DB, Helmers KF, Jobe JB, Leveck MD, Marsiske M, 178–81.
et al.Effects of cognitive training interventions with older adults: a Gratzinger 1990 {published data only}
randomized controlled trial. JAMA 2002;288(18):2271–81. Gratzinger P, Sheikh JI, Friedman L, Yesavage JA. Cognitive
Buiza 2007 {published data only} interventions to improve face-name recall: The role of personality
Buiza C, Etxeberria I, Galdona N, González MF, Arriola E, López trait differences. Developmental Psychology 1990;26(6):889–93.
de Munain A, et al.A randomized, two-year study of the efficacy of Hill 1987 {published data only}
cognitive intervention on elderly people: the Donostia Hill RD, Sheikh JI, Yesavage J. The effect of mnemonic training on
Longitudinal Study. International Journal of Geriatric Psychiatry perceived recall confidence in the elderly. Experimental Aging
2007;23(1):85–94. [DOI: 10.1002/gps.1846] Research 1987;13(4):185–8.
Caprio 1996 {published data only} Hill 1988 {published data only}
Caprio-Prevette, M. D, Fry. P. S. Memory enhancement program Hill RD, Sheikh JI, Yesavage J. Pretraining enhances mnemonic
for community-based older adults: Development and evaluation. training in elderly adults. Experimental Aging Research 1988;14(4):
Experimental Aging Research 1996;22:281–303. 207–11.
Craik 2007 {published data only} Hill 1990 {published data only}
Craik FIM, Winocur G, Palmer H, Binns MA, Edwards M, Bridges Hill RD, Storandt M, Simeone C. The effects of memory skills
K, et al.Cognitive rehabilitation in the elderly: Effects on memory. training and incentives on free recall in older learners. Journal of
Journal of the International Neuropsychological Society 2007;13: Gerontology: Psychological Sciences 1990;45:227–232.
132–42.
Hill 1991 {published data only}
De Vreese 1996 {published data only} Hill RD, Allen C, McWhorter P. Stories as a mnemonic aid for
De Vreese LP, Neri M, Boiardi R, Ferrari P, Belloi L, Salvioli G. older learners. Psychology and Aging 1991;6:484–6.
Memory training and drug therapy act differently on memory and
metamemory functioning: Evidence from a pilot study. Archives of Levine 2007 {published data only}
Gerontology and Geriatrics 1996;Suppl 5:9–22. Levine B, Stuss DT, Winocur G, Binns MA, Fahy L, Mandic M, et
al.Cognitive rehabilitation in the elderly: Effects on strategic
Derwinger 2005 {published data only} behavior in relation to goal management. Journal of the
Derwinger A, Stigsdotter Neely A, MacDonald S, Backman L. International Neuropsychological Society 2007;13:143–52.
Forgetting numbers in old age: strategy and learning speed matter.
Gerontology 2005;51(4):277–84. Mahncke 2006 {published data only}
Mahncke HW, Connor BB, Appelman J, Ahsanuddin ON, Hardy
Dunlosky 2007 study 2 {published data only} J, Wood RA, et al.Memory enhancement in healthy older adults
Dunlosky J, Cavallini E, Roth H, McGuire CL, Vecchi T, Hertzog using a brain plasticity-based training program: a randomized,
C. Do self-monitoring interventions improve older adult learning?. controlled study. Proceedings of the National Academy of Sciences
Journal of Gerontology 2007;62B(Special Issue):70–76. 2006;103(33):12523–8.
Dunlosky 2007 study1 {published data only} Margrett 2006 {published data only}
Dunlosky J, Cavallini E, Roth H, McGuire CL, Vecchi T, Hertzog Margrett JA, Willis SL. In-home cognitive training with older
C. Do self-monitoring interventions improve older adult learning?. married couples: Individual versus collaborative learning. Aging,
Journal of Gerontology 2007;62B(Special Issue):70–76. Neurpsychology, and Cognition 2006;13:173–95.
Edwards 2005 {published data only} Piccolini 1992 {published data only}
Edwards JD, Wadley VG, Vance DE, Wood K, Roenker DL, Ball Piccolini C, Amadio L, Spazzafumo L, Moroni S, Freddi A. The
KK. The impact of speed of processing training on cognitive and effects of a rehabilitation program with mnemotechniques on
everyday performance. Aging and Mental Health 2005;9(3):262–71. institutionalized elderly subjects. Archives of Gerontology and
Geriatrics 1992;15:141–149.
Fabre 2002 {published data only}
Fabre C, Chamari K, Mucci P, Massé-Biron J, Préfaut C. Rapp 2002 {published data only}
Improvement of cognitive function by mental and/or Rapp S, Brenes G, Marsh AP. Memory enhancement training for
individualized aerobic training in healthy elderly subjects. older adults with mild cognitive impairment: a preliminary study.
International Journal of Sports Medicine 2002;23:415–25. Aging & Mental Health 2002;6(1):5–11.
Rasmusson 1999 {published data only} Anschutz 1985 {published data only}
Rasmusson DX, Rebok GW, Bylsma FW, Brandt J. Effects of three Anschutz L, Camp CJ, Markley RP, Kramer JJ. Maintenance and
types of memory training in normal elderly. Aging, Neuropsychology generalization of mnemonics for grocery shopping by older adults.
and Cognition 1999;6:56–66. Experimental Aging Research 1985;11(3):157–60.
Rozzini 2007 {published data only} Anschutz 1987 {published data only}
Rozzini, L, Costardi, D, Vicini Chilovi, B, Franzoni, S, Trabucchi, Anschutz, L, Camp, C. J, Markley, R. P, & Kramer, J. J.
M, Padovani, A. Efficacy of cognitive rehabilitation in patients with Remembering mnemonics: a three-year follow-up on the effects of
mild cognitive impairment treated with cholinesterase inhibitors. mnemonics training in elderly adults. Experimental Aging Research
International Journal of Geriatric Psychiatry 2007;22(4):356–360. 1987;13(3):141–143. [DOI: 10.1080/03610738708259315]
[DOI: 10.1002/gps.1681] Baldelli 1991 {published data only}
Schaffer 1992 {published data only} Baldelli, M.V, Toschi, A, Ayyud, N, Spanó, A, Andermarcher E.
Schaffer G, Poon L. Individual variability in memory training with The elderly and memory training: No differences were encountered
the elderly. Educational Gerontology 1982;8:217–29. between the two sexes. Archives of Gerontology and Geriatrics 1991;
Supplement(2):147–150.
Scogin 1985 {published data only}
Scogin F, Storandt M, Lott L. Memory-skills training, memory Baltes 1988 {published data only}
complaints, and depression in older adults. Journal of Gerontology Baltes PB, Kliegl R, Dittmann-Kohli F. On the locus of training
1985;40:562–8. gains in research on the plasticity of fluid intelligence in old age.
Journal of Educational Psychology 1988;80(3):392–400.
Scogin 1992 {published data only}
Scogin, Prohaska, Weeks. The efficacy of self-taught memory Baltes 1989 {published data only}
training for community-dwelling older adults. Educational Baltes, P.B, Sowarka, D, Kliegl. R. Cognitive training research on
Gerontology 1992;18:751–66. fluid intelligence in old age: What can older adults achieve by
themselves?. Psychology and Aging 1989;4(2):217–221.
Stigsdotter 1989 {published data only}
Stigsdotter A, Bäckman L. Multifactorial memory training with Belleville 2006 {published data only}
older adults: How to foster maintenance of improved performance. Belleville, S, Gilbert, B, Fontaine, F, Gagnon, L, Ménard, E,
Gerontology 1989;35:260–7. Gauthier. S. Improvement of episodic memory in personsn with
mild cognitive impairment and healthy oder adults: evidence from
Stigsdotter 1993 study 1 {published data only}
a cognitive intervention program. Dementia and Geriatric Cognitive
Stigsdotter NA, Bäckman L. Long-term maintenance of gains from
Disorders 2006;22:486–499.
memory training in in older adults: Two 3.5-year follow-up studies.
Journal of Gerontolgy: Psychological Sciences 1993;48(5):233–237. Best 1992 {published data only}
Best, Hamlett, Davis. Memory complaint and memory
Stigsdotter 1993 study 2 {published data only} performance in the elderly: the effects of memory skills training
Stigsdotter NA, Bäckman L. Long-term maintenance of gains from and expectancy change. Applied Cognitive Psychology 1992;6(5):
memory training in older adults: Two 31/2-year follow-up studies. 405–416. [DOI: 10.1002/acp.2350060505]
Journal of Gerontology: Psychological Sciences 1993;48(5):233–237.
Bond 2000 {published data only}
Stigsdotter 1995 {published data only} Bond, G.E, Wolf-Wilets, V, Fiedler, F.E, & Burr, R.L. Computer-
Stigsdotter NA, Bäckman L. Effects of multifactorial memory aided cognitive training of the aged: a pilot study. Clinical
training in old age: generalizability across tasks and individuals. Gerontologist 2000;22(2):19–42.
Journal of Gerontology 1995;50B(3):134–40.
Boron 2007 {published data only}
Valentijn 2005 {published data only} Boron, J.B, Turiano, N.A, Willis, S.L, Schaie. W.K. Effects on
Valentijn SAM, van Hooren SAH, Bosma H, Touw DM, Jolles J, cognitive training on change in accuracy in inductive reasoning
van Boxtel MPJ, et al.The effect of two types of memory training ability. Journal of Gerontology: Psychological Sciences 2007;62B(3):
on subjective and objective memory performance in healthy 179–186.
individuals aged 55 years and older: a randomized controlled trial. Boron, J.B, Willis, S.L, Schaie. K.W. Cognitive training gain as a
Patient Education and Counselling 2005;57:106–14. predictor of mental status. Journal of Gerontology: Psychological
Yesavage 1990 {published data only} Sciences 2007;62B(1):45–52.
Yesavage JA, Sheikh JI, Friedman L, Tanke E. Learning mnemonics: Brooks 1993 {published data only}
roles of aging and subtle cognitive impairment. Psychol Aging 1990; Brooks, J.O. 3rd, Friedman, L, Yesavage, J.A. A study of the
5:133–7. problems older adults encounter when using a mnemonic
technique. International Psychogeriatrics 1993;5(1):57–65. [DOI:
References to studies excluded from this review
10.1017/S1041610293001395]
Andrewes 1996 {published data only} Brooks 1999 {published data only}
Andrewes DG, Kinsella G, Murphy M. Using a memory handbook Brooks, J.O. III, Friedman, L, Pearman, A.M, Yesavage. J.A.
to improve everyday memory in community-dwelling older adults Mnemonic training in older adults: effects of age, length of
with memory complaints. Experimental Aging Research 1996;22: training, and type of cognitive retraining. International
305–22. Psychogeriatrics 1999;11(1):75–84.
Calero 1997 {published data only} 20, 135-148. Training effects on intelligence of older persons.
Calero MD, Garcia-Berbén TM. A self-training program in Archives of Gerontology and Geriatrics 1995;20(2):135–148. [DOI:
inductive reasoning for low-education elderly: Tutor-guided 10.1016/0167-4943(94)00591-T]
training versus self-training. Archives of Gerontology and Geriatrics
Finkel 1989 {published data only}
1997;24:249–59.
Finkel, S.I, Yesavage. J.A. Learning mnemonic: A preliminary
Cavallini 2002 {published data only} evaluation of a computer-aided instruction package for the elderly.
Cavallini, E, Pagnin, A, & Vecchi, T. The rehabilitation of memory Experimental Aging Research 1989;15(3-4):199–201. [DOI:
in old age: effects of mnemonics and metacognition in strategic 10.1074/jbc.M202849200]
training. Clinical Gerontologist 2002;26(1-2):125–141. [DOI: doi:
10.1016/j.pec.2006.07.010] Fleischmann 1985 {published data only}
Fleischmann, U.M. Gedächtnisbezogene Förderung im hohen
Cavallini 2003 {published data only} Lebensalter. Die Rehabilitation 1985;24(1):36–38.
Cavallini, E, Pagnin, A, Vecchi. T. Aging and everyday memory:
the beneficial effect of memory training. Archives of Gerontology and Gunther 2003 {published data only}
Geriatrics 2003;37(3):241–257. [DOI: 10.1016/S0167-4943 Günther VK, Schäfer P, Holzner BJ, Kemmler GW. Long-term
(03)00063-3] improvements in cognitive performance through computer-assisted
cognitive training: a pilot study in a residential home for older
Cerella 2006 {published data only}
people. Aging & Mental Health 2003;7:200–6.
Cerella, J, Onyper, S.V, Hoyer. W.J. The associative-memory basis
of cognitive skill learning: Adult age differences. Psychology and Hill 1989 {published data only}
Aging 2006;21(3):483–498. Hill, R.D, Yesavage, J.A, Sheikh, J, & Friedman, L. Mental status as
Cipriani 2006 {published data only} a predictor of response memory training in older adults.
Cipriani, G, Bianchetti, A, & Trabucchi, M. Outcomes of a Educational Gerontology 1989;15(6):633–639. [DOI: 10.1080/
computer-based cognitive rehabilitation program on Alzheimer’s 0380127890150607]
disease patients compared with those on patients affected by mild Israel 1998 {published data only}
cognitive impairment. Archives of Gerontology and Geriatrics 2006; Israel, L, Myslinski, M, Kozarevic. D. Nootropic treatment and
43(3):327–333. [DOI: 10.1016/j.archger.2005.12.003] combined therapy in age-associated memory impairment. Archives
De Vreese 1998 {published data only} of Gerontology and Geriatrics 1998;26(S1):269–274. [DOI:
De Vreese, L. P, Belloi, L, Iacono, S, Finelli, C, Neri, M. Memory 10.1016/S0167-4943(98)80038-1]
training programs in memory complainers: Efficacy on objective Johnston 1989 {published data only}
and subjective memory functioning.. Archives of Gerontology and Johnston, L, Gueldner. S.H. Remember when...? Using
Geriatrics. 1998;Suppl. 6:141–154. mnemonics to boost memory in the elderly. Journal of
Derwinger 2003 {published data only} Gerontological Nursing 1989;15(8):22–26.
Derwinger, A, Stigsdotter Neely, A, Persson, M, Hill, R.D, &
Kramer 1995 {published data only}
Backman L. (2003). . Aging, Neuropsychology, Cognition. 10,
Kramer, A.R, Larish, J.F, Strayer. D.L. Training for attentional
202-214. Remembering numbers in old age: mnemonic training
control in dual task settings: A comparison of young and old adults.
versus self-generated strategy training. Aging, Neuropsychology and
Journal of Experimental Psychology: Applied 1995;1(1):50–76.
Cognition 2003;10(3):202–214.
Labouvie-Vief 1976 {published data only}
Derwinger 2005 b {published data only}
Labouvie-Vief, G, Gonda. J.N. Cognitive strategy training and
Derwinger S, Stigsdotter NA, Bäckman L. Design your own
intellectual performance in the elderly. Journal of Gerontology 1976;
memory strategies! Self-generated strategy training versus
31(3):327–332. [DOI: 10.1093/geronj/31.3.327]
mnemonic training in old age: an 8 month follow-up.
Neuropsychological Rehabilitation 2005;15(1):37–54. [DOI: Lachman 1992 {published data only}
10.1080/09602010343000336] Lachman, Weaver, Bandura, Elliot, Lewkowicz. Improving
Dittman-Kohli 1991 {published data only} memory and control beliefs through cognitive restructuring and
Dittmann-Kohli, F, Lachman, M.E, Kliegl, R, Baltes. P.B. Effects self-generated strategies. Journal of Gerontology: Psychological
of cognitive training and testing on intellectual efficacy beliefs in Sciences 1992;47(5):293–299.
elderly adults. Journal of Gerontology 1991;46(4):162–164. Martin 1998 {published data only}
Erickson 2007 {published data only} Martin, M, Kayser. N. (1998), 31, 91-103. Das modulare
Erickson, K.I, Colcombe, S.J, Wadhwa, R, Bherer, L, Peterson, Gedächtnistraining für ältere Erwachsene: Konzeption und
M.S, Scalf, P.E, Kim, J.S, Alvarado, M, Kramer, A.F. Training- Erprobung. Zeitschrift für Gerontologie und Geriatrie 1998;31(1):
induced plasticity in older adults: Effects of training on 91–103.
hemispheric asymmetry. Neurobiology of Aging 2007;28(2): McKitrick 1999 {published data only}
272–283. [DOI: 10.1016/j.neurobiolaging.2005.12.012] McKitrick, L.A, Friedman, L.F, Brooks, J.O. 3rd, Pearman, A,
Fernandez 2005 {published data only} Kraemer, H.C, Yesavage, J.A. Predicting response of older adults to
Fernández-Ballesteros, R, Calero. M.D. (1995). Training effects on mnemonic training: who will benefit?. International Psychogeriatrics
intelligence of older persons. Archives of Gerontology, Geriatrics. 1999;11(3):289–300. [DOI: 10.1017/S1041610299005852]
Mohs 1998 {published data only} Panza 1996 {published data only}
Mohs, R.C, Ashman, T.A, Jantzen, K, Albert, M, Brandt, J, Panza, F, Solfrizzi, V, Mastroianni, F, Nardo, G.A, Cigliola, F, &
Gordon, B, Rasmusson, X, Grossman, M, Jacobs, D, Stern. Y. A Capurso, A. A rehabilitation program for mild memory
study of the efficacy of a comprehensive memory enhancement impairments. Archives of Gerontology and Geriatrics 1996;Suppl. 5:
program in healthy elderly persons. Psychiatry Research 1998;77(3): 51–55.
183–195. [DOI: 10.1016/S0165-1781(98)00003-1] Paxton 2006 {published data only}
Neils-Strunjas 1997 {published data only} Paxton, J.L, Barch, D.M, Storandt, M, Braver. T.S. Effects of
Neils-Strunjas, J, Ollier, L, Thomas, K, Thomas. G. A comparison environmental support and strategy training on older adults’ use of
of two training methods for teaching name-face associations to context. Psychology and Aging 2006;21(3):499–509.
elderly subjects with self-reported memory loss. Brain and Rebok 1989 {published data only}
Cognition 1997;35:308–426. Rebok, G.W, Balcerak. L.J. Memory self-efficacy and performance
Olazarán 2004 {published data only} differences in young and old adults: the effect of mnemonic
Olazarán, J, Muñiz, R, Reisberg, B, Peña-Casanova, J, del Ser, T, training. Developmental Psychology 1989;25(5):714–721.
Cruz-Jentoft, A.J, &, et al.Benefits of cognitive-motor intervention Rebok 1996 {published data only}
in MCI and mild to moderate Alzheimer disease. Neurology 2004; Rebok, G.W, Rasmusson, D.X, Brandt. J. Prospects for
63:2348–2353. computerized memory training in normal elderly: Effects of
Onyper 2006 {published data only} practice on explicit and implicit memory tasks. Applied Cognitive
Onyper SV, Hoyer WJ, Cerella J. Determinants of retrieval Psychology 1996;10(3):211–223. [DOI: 10.1002/acp.2350100301]
solutions during cognitive skill training: Source confusions. Rebok 1997 {published data only}
Memory and Cognition 2006;34(3):538–549. Rebok, G.W, Rasmusson, D.X, Bylsma, F.W, & Brandt, J. Memory
Oswald 1996 {published data only} improvement tapes: how effective for elderly adults?. Aging,
Oswald, W.D, Rupprecht, R, Gunzelmann, T, Tritt. K. The SIMA- Neuropsychology and Cognition 1997;4(4):304–311. [DOI:
project: Effects of 1 year cognitive and psychomotor training on 10.1080/13825589708256655]
cognitive abilities of the elderly. Behavioural Brain Research 1996; Saczynski 1992 {published data only}
78(1):62–72. [DOI: 10.1016/0166-4328(95)00219-7] Saczynski, J.S, Willis, S.L, Schaie K.W. Strategy use in reasoning
training with older adults. Aging, Neuropsychology and Cognition
Oswald 1998 {published data only}
Oswald, W.D, Hagen, B, Rupprecht. R. Bedingungen der 2002;9(1):48–60.
Erhaltung und Förderung von Selbständigkeit im höheren Schaie 1987 {published data only}
Lebensalter (SIMA) - Teil X: Verlaufsanalyse des kognitiven Status. Schaie, K.W, Willis, S.L, Hertzog, Ch, Schulenberg. J.E. Effects of
Zeitschrift für Gerontopsychologie und -psychiatrie 1998;11(3): cognitive training on primary mental ability structure. Psychology
202–221. and Aging 1987;2(3):233–242.
Oswald 2002 {published data only} Schmidt 1999 {published data only}
Oswald, W.D, Hagen, B, Rupprecht, R, Gunzelmann. T. (2002). Schmidt IW, Dijkstra HT, Berg IJ, Deelman BG. Memory training
Conditions of conservation and promoting independence in old- for remembering names in older adults. Clinical Gerontologist 1999;
age (SIMA) Part XVII: Summary the long-term training effects 20(2):57–73.
[Bedingungen der Erhaltung und Förderung von Selbständigkeit Schmidt 2001 {published data only}
im höheren Lebensalter (SIMA). Teil XVII: Zusammenfassende Schmidt IW, Berg, IJ, Deelman BG. Prospective memory training
Darstellung der langfristigen Trainingseffekte]. Zeitschrift für in older adults. Educational Gerontology 2001;27(6):455–78.
Gerontopsychologie und - psychiatrie 2002;15(1):13–31.
Schmitt 1981 {published data only}
Oswald 2006 {published data only} Schmitt F.A, Murphy M.D, Sanders R.E. Training older adult free
Oswald, W.D, Gunzelmann, T, Rupprecht, R, Hagen. B. recall rehearsal strategies. Journal of Gerontology 1981;36(3):
Differential effects of single versus combined cognitive and physical 329–337.
training with older adults: the SimA study in a 5-year perspective.. Scogin 1988 {published data only}
European Journal of Aging 2006;3(4):179–192. [DOI: 10.1007/ Scogin, F, Bienias. J.L. A three-year follow-up of older adult
s10433-006-0035-z] participants in a memory-skills training program. Psychology and
Oswald 2007 {published data only} Aging 1988;3(4):334–337.
Oswald, W.D, Gunzelmann, T, Ackermann. A. Effects of a Sheikh 1986 {published data only}
multimodal activation program (SimA-P) in residents of nursing Sheikh, J.I, Hill. R.D, Yesavage. J.A. Long-term efficacy of
homes. European Review of Aging and Physical Activity 2007;4(2): cognitive training for age-associated memory impairment: A six-
91–102. [DOI: 10.1007/s11556-007-0025-y] month follow-up study. Developmental Neuropsychology 1986;2(4):
Ott-Chervet 1998 {published data only} 413–421.
Ott-Chervet, C, Rüegger-Frey, B, Klaghofer, R, & Six, P. Sitzer 2006 {published data only}
Evaluation eines computergestützten kognitiven Trainings mit Sitzer, DI, Twamley EW, & Jeste DV. Cognitive training in
hochbetagten Patienten eines geriatrischen Krankenhauses. Alzheimer’s disease: a meta-analysis of the literature. Acta
Zeitschrift für Gerontopsychologie und -psychiatrie 1998;1(1):13–23. Psychiatrica Scandinavica 2006;114:75–90.
Small 2006 {published data only} West 1992 {published data only}
Small, G.W, Silverman, D.H, Siddarth, P, Ercoli, L.M, Miller, K.J, West, R. L, Crook. T. H. Video training of imagery for mature
Lavretsky, H, Wright, B.C, Bookheimer, S.Y, Barrio, J.R, Phelps, adults. Applied Cognitive Psychology 1992;6(4):307–320. [DOI:
M.E. Effects of a 14-day healthy longevity lifestyle program on 10.1002/acp.2350060404]
cognition and brain function. American Journal of Geriatrics and
Willis 1986 {published data only}
Psychiatry 2006;14(6):538–545. [DOI: 10.1097/
Willis, S.L, Schaie. K.W. Training the elderly on the ability factors
01.JGP.0000219279.72210.ca]
of spatial orientation and inductive reasoning. Psychology and Aging
Stine-Morrow 2007 {published data only} 1986;1(3):239–247.
Stine-Morrow, E.A.L, Parisi, J.M, Morrow, D.G, Greene, J, Park.
D.C. An engagement model of cognitive optimization through
Willis, S.L, Schaie. K.W. Gender differences in spatial ability in old
adulthood. Journal of Gerontology 2007;62B(SI1):62–69.
age: Longitudinal and intervention findings. Sex Roles 1988;81(3-
Talassi 2007 {published data only} 4):189–203.
Talassi, E, Guerreschi, M, Feriani, F, Fedi, V, Bianchetti, A, &
Trabucchi, M. Effectiveness of a cognitive rehabilitation program in
Willis, S.L, Nesselroade. C.S. Long-term effects of fluid ability
mild dementia (MD) and mild cognitive impairment (MCI) a case
training in old-age. Developmental Psychology 1990;26(6):905–910.
control study. Archives of Gerontology and Geriatrics 2007;Suppl. 1:
[DOI: 10.1037/0012-1649.26.6.905]
391–399.
Valenzuela 2003 {published data only} Willis 2006 {published data only}
Valenzuela, M.J, Jones, M, Wen, W, Rae, C, Graham, S, Shnier, R, Willis, S.L, Tennstedt, S.L, Marsiske, M, Ball, K, Elias, J, Mann
Sachdev. P. Memory training alters hippocampal neurochemistry in Koepke, K, Morris, J.N, Rebok, G.W, Unverzagt, F.W, Stoddard,
healthy elderly. Neuroreport 2003;14(10):1333–1337. A.M, Wright. E. Long-term effects of cognitive training on
everyday functional outcomes in older adults. Journal of the
Van Gerven 2003 {published data only} American Medical Association 2006;296(23):2805–2814.
Van Gerven, P.W.M, Paas, F, van Merrienboer, J.J.G, Hendricks,
M, Schmidt. H.G. The efficacy of multimedia learning into old Winningham 2003 {published data only}
age. British Journal of Educational Psychology 2003;73(4):489–505. Winningham, R.G, Anunsen, R, Hanson, L.M, Laux, L, Kaus,
[DOI: 10.1348/000709903322591208] K.D, & Reifers, A. MemAerobics: A cognitive intervention to
improve memory ability and reduce depression in older adults.
Van Gerven 2006 {published data only} Journal of Mental Health and Aging 2003;9(2):182–192.
Van Gerven, P.W.M, Paas, F, van Merrienboer, J.J.G, Schmidt.
H.G. Modality and variability as factors in training the elderly. Winocur 2007 {published data only}
Applied Cognitive Psychology 2006;20(3):311–320. [DOI: 10.1002/ Winocur, G, Palmer, H, Dawson, D, Binns, M.A, Bridges, K,
acp.1247] Stuss. D.T. Cognitive rehabilitation in the elderly: An evaluation of
psychosocial factors. Journal of the International Neuropsychological
Van Hooren 2007 {published data only}
Society 2007;13(1):153–165. [DOI: 10.1017/
Van Hooren, S.A.H, Valentijn, S.A.M, Bosma, H, Ponds, S135561770707018X]
R.W.H.M, Boxtel, M.P.J, Levine, B, Robertson, I.& Jolles, J. Effect
of a structured course involving goal management training in older Wolinksy 2006 b {published data only}
adults: A randomised controlled trial. Patient Education and Wolinsky, F.D, Unverzagt, F.W, Smith, D.M, Jones, R, Stoddard,
Counseling 2007;65(2):205–213. [DOI: 10.1016/ A, Tennstedt. S.L. (2006), 61A(12), 1324-1329. The ACTIVE
j.pec.2006.07.010] cognitive training trial and health-related quality of life: protection
that lasts for 5 years. The Journals of Gerontology Series A: Biological
Wadley 2006 {published data only} Sciences and Medical Sciences 2006;61A(12):1324–1329.
Wadley, V.G, Benz, R.L, Ball, K.K, Roender, D.L, Edwards, J.D,
Vance. D.E. Development and evaluation of home-based speed-of- Wolinsky 2006 a {published data only}
processing training for older adults. Archives of Physical Medecine Wolinsky, F.D, Unverzagt, F.W, Smith, D.M, Jones, R, Stoddard,
and Rehabilitation 2006;87(6):757–763. [DOI: 10.1016/ A, Tennstedt. S.L. The effects of the ACTIVE cognitive training
j.apmr.2006.02.027] trial on clinically relevant declines in health-related quality of life.
Journals of Gerontology: Social Sciences 2006;61B(5):281–287.
Wenisch 2007 {published data only}
Wenisch E, Cantegreil-Kalen I, De Rotrou J, Garrigue P, Moulin F, Wood 1987 {published data only}
Batouche F, et al.Cognitive stimulaton intervention for elders with Wood, L.E, Pratt. J.D. Pegword mnemonic as an aid to memory in
mild cognitive impairment compared with normal aged subjects the elderly: a comparison of four age groups. Educational
preliminary results.. Aging Clinical and Experimental Research 2007; Gerontology 1987;13(4):325–339.
19(4):316–322. Yesavage 1983 a {published data only}
Werner 2000 {published data only} Yesavage, J.A, Rose, T.L, Bower. G.H. Interactive imagery and
Werner P. Assessing the effectiveness of a memory club for elderly affective judgement improve face-name learning in the elderly..
persons suffering from mild cognitive impairment. Clinical Journal of Gerontology 1983;38(2):197–203. [DOI: 10.1093/
Gerontologist 2000;22(1):3–14. geronj/38.2.197]
Yesavage 1983 b {published data only} Hultsch 1999
Yesavage, J.A. Imagery pretraining and memory training in the Hultsch DF, Hertzog C, Small BJ, Dixon RA. Use it or lose it:
elderly. Gerontology 1983;29(4):271–275. [DOI: DOI:10.1159/ engaged lifestyle as a buffer of cognitive decline in aging?.
000213126] Psychology and Aging 1999;14:245–263.
Yesavage 1983 c {published data only} Jungwirth 2005
Yesavage, J.A, Rose. T.L. Concentration and mnemonic training in Jungwirth S, Weissgram S, Zehetmayer S, et al.VITA: subtypes of
elderly subjects with memory complaints: a study of combined mild cognitive impairment in a community-based cohort at the age
therapy and order effects. Psychiatry Research 1983;9(2):157–167. of 75 years. International Journal of Geriatric Psychiatry 2005;20:
[DOI: 10.1016/0165-1781(83)90037-9] 452–8.
Yesavage 1984 {published data only} Kliegl 1989
Yesavage, J. A, & Jacob, R. Effects of relaxation and mnemonics on Kliegl R, Smith J, Baltes PB. Testing-the-Limits and the Study of
memory, attention and anxiety in the elderly. Experimental Aging Adult Age-Differences in Cognitive Plasticity of a Mnemonic Skill.
Research 1984;10(4):211–214. [DOI: 10.1080/ Developmental Psychology 1989;25(2):247–56.
03610738408258467]
Kral 1962
Zarit 1981 {published data only}
Kral VA. [Senescent forgetfulness: Benign and malignant]. Le
Zarit, S.H, Cole, K.D, Guider. R.L. Memory training strategies
Journal de l’Association Médicale Canadienne 1962;86:257–60.
and subjective complaints of memory in the aged. The Gerontologist
1981;21(2):158–164. [DOI: 10.1093/geront/21.2.158] Kratz 2002
Kratz B. Drei Aspekte des diagnostischen Konzepts der leichten
Additional references kognitiven Beeinträchtigung im Alter. Dissertationsschrift,
APA 1987 University of Heidelberg 2002.
American Psychiatric Association. Diagnostic and Statistical Manual Kumar 2005
of Mental Disorders. 3rd Edition. Washington, DC: American Kumar R, Dear KBG, Christensen H, Ilschner S, Jorm AF, Meslin
Psychiatric Association, 1987. C, Rosenman SJ, Sachdev PS. Prevalence of mild cognitive
Busse 2003 impairment in 60-to 64-year-old community-dwelling individuals:
Busse A, Bischkopf J, Riedel-Heller SG, et al.Mild cognitive The Personality and Total Health through Life 60+study. Dementia
impairment: prevalence and predictive validity according to current and Geriatric Cognitive Disorders 2005;19:67–74.
approaches. Acta Neurologica Scandinavia 2003;108:71–81. Larrieu 2002
Chalmers 1983 Larrieu S, Letenneur L, Orgogozo JM, Fabrigoule C, Amieva H, Le
Chalmers TC, Celano P, Sacks HS, Smith H. Bais in treatment Carret N, Barberger-Gateau P, Dartigues JF. Incidence and
assignment in controlled clinical trials. New England Journal of outcome of mild cognitive impairment in a population-based
Medicine 1983;309:1358–61. prospective cohort. Neurology 2002;59:1594–9.
Clare 2003 Manly 2005
Clare L. Cognitive training and cognitive rehabilitation for people Manly JJ, Bell-McGinty S, Tang MX, et al.Implementing diagnostic
with early-stage dementia. Reviews in Clinical Gerontology 2003;13: criteria and estimating frequency of mild cognitive impairment in
75–83. an urban community. Archives of Neurology 2005;62:1739–46.
Crook 1986
Nyberg 2005
Crook T, Bartus RT, Ferris SH, Whitehouse P, Cohen GD,
Nyberg L. Cognitive training in healthy aging: A cognitive
Gershon S. Age-associated memory impairment: proposed
neuroscience perspective. In: Cabeza R, Nyberg L, Park DC editor
diagnostic criteria and measures of clinical change. Developmental
(s). Cognitive neuroscience of aging. New York: Oxford University
Neuropsychology 1986;2:261–76.
Press, 2005:309–21.
Graham 1997
Petersen 2001
Graham JE, Rockwood K, Beattie BL, Eastwood R, Gauthier S,
Petersen RC, Doody R, Kurz A, Mohs RC, Morris JC, Rabins PV,
Tuokko H, McDowell I. Prevalence and severity of cognitive
Ritchie K, Rossor M, Thal L, Winblad B. Current concepts in Mild
impairment with and without dementia in an elderly population.
Cognitive Impairment. Archives of Neurology 2001;58:1985–92.
Lancet 1997;349:1793–96.
Higgins 2008 Purser 2005
Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Purser JL, Fillenbaum GG, Pieper CF, et al.Mild cognitive
Reviews of Interventions Version 5.0.0 [updated February 2008]. impairment and 10-year trajectories of disability in the Iowa
The Cochrane Collaboration, 2008. Available from www.cochrane- established populations for epidemiologic studies of the elderly
handbook.org. Chichester, UK: John Wiley & Sons, Ltd. cohort. Journal of the American Geriatrics Society 2005;53:1966–72.
Hoyer 2006 Schooler 2001
Hoyer WJ, Verhaeghen P. Memory aging. In: Birren JE, Schaie Schooler C, Mulatu MS. The reciprocal effects of leisure time
KW editor(s). Handbook of the psychology of aging. 6th Edition. activities and intellectual functioning in older people: a
Amsterdam: Academic Press, 2006:209–32. longitudinal analysis. Psychology and Aging 2001;16:466–82.
Schulz 1995 Verhaeghen 1992
Schulz KF, Chalmers I, Hayes RJ, Altman D. Empirical evidence of Verhaeghen P, Marcoen A, Goossens L. Improving memory
bias. Journal of the American Medical Association 1995;273:408–12. performance in the aged through mnemonic training: A meta-
Singer 2003 analytic study. Psychology and Aging 1992;7:242–51.
Singer T, Lindenberger U, Baltes PB. Plasticity of memory for new
learning in very old age: A story of major loss?. Psychology and Willis 2001
Aging 2003;18:306–17. Willis SL. Methodological issues in behavioural intervention
research with the elderly. In: Birren JE, Schaie KW editor(s).
Stern 2002
Handbook of the psychology of aging. 5th Edition. San Diego, CA:
Stern Y. What is cognitive reserve? Theory and research application
Academic Press, 2001:78–108.
of the reserve concept. Journal of the International
Neuropsychological Society 2002;8:448–60. Wilson 2002
Unverzagt 2007 Wilson RS, Mendes de Leon CF, Barnes LL, et al.Participation in
Unverzagt FW, Kasten L, Johnson KE, Rebok GW, Marsiske M, cognitively stimulating activities and risk of incident Alzheimer
Mann Koepke K, Elias JW, et al.Effect of memory impairment on disease. Journal of the American Medical Association 2002;287:
training outcomes in ACTIVE. Journal of the International 742–8.
Neuropsychological Society 2007;13(6):953–960. ∗
Indicates the major publication for the study
CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]
Ball 2002
Methods single-blind design; no contact control group;training lasts 5-6 weeks
Participants - n no contact control=704

- n memory=711
- n problem=705
Interventions problem-solving training, memory strategies, speed training
Outcomes problem solving, verbal episodic memory, speed;
Notes
Buiza 2007
Methods double-blind design; 2 years (total of 180 sessions), t1=baseline, then every 6 months
Participants - n exp structured 1=85

- n exp unstructured 2=68
Interventions attention, orientation, memory, language, visuoconstruction, visuomanual coordination, praxis
Outcomes Luria, speed (TMT), visuomanual coordination, short term memory, immediate recall, recent logic execution memory,
abstraction proverbs, phonematic fluency, IADL
Notes data available: initial, annual, biannual
Caprio 1996
Methods memory training versus alternative training; no no contact control group
Participants age range 65-76 years
Interventions cognitive restructuration technique versus traditional memory training
Outcomes Guild Memory Test, supermarket test; subjective memory tests and Geriatric Depression Scale
Notes
Craik 2007
Methods cross-over design; early training group vs. late training; group (before cross-over); baseline, post-test after 3 months;
4 weeks duration
Participants n early =29

n late(control)=20
Interventions memory skills training
Outcomes alpha span (working memory) total score,

Brown-Peterson: secondary memory,
primary memory, 0s delay,
primary memory ,3s delay
Notes
De Vreese 1996
Methods 4 groups:
- memory training
- drug treatment
- drug + memory
- no contact control
Participants MCI patients: a score > 25 adjusted for age and schooling (Measso et al., 1993) on the mini-mental state examination
(MMSE, Folstein et al., 1975); (f ) no clinically relevant depression as disclosed by a score < 16 on the geriatric
depression scale (GDS, Yesavage et al., 1983); (g) presence of impaired objective memory resulting in a score < 15.76
on the story recall test (De Renzi, 1977) and/or significant memory complaints evinced by a score > 20 on the
cognitive difficulties scale (CDS, MacNair and Kahn, 1983).
n memory training=10
n drug treatment=7
n memory training and drug treatment=10
n control group=8
Interventions drug treatment and memory training
Outcomes Randt Memory Test:acquisition, delayed recall, memory index
Notes
Derwinger 2005
Methods 2 experimental groups and 1 no contact control; 5 weeks, group setting
Participants n =20/group
Interventions number-consonant mnemonic versus self-generated strategy training versus control group
Outcomes free recall of digit spans
Derwinger 2005 (Continued)
Notes
Dunlosky 2007 study 2
Methods 2x2h training sessions with pause of 2 weeks in paired associate learning for 2 groups, 1 CG; pre-post-design
Participants n strategy/imagery training=34

n self-monitoring training=38
n control=29
Interventions strategy/imagery vs. self-monitoring training and no contact control
Outcomes correctly recalled word-pairs: all training groups better than control group, but significant differences between groups
Notes
Dunlosky 2007 study1
Methods 2 sessions - length?

4 groups:
Participants n strat/imag=21
n self-monit.=21
n comb=23
n control=20
Interventions self-monitoring approach for improving older adult learning
Outcomes correctly recalled word-pairs: no sign. differences between training groups and control
Notes
Edwards 2005
Methods 5 weeks
Participants n speed=63
n internet-training control =63
Interventions speed of processing training
Outcomes cognitive performance and IADL: UFOV,

Road Sign Test,
timed IADL,
letter comparison
Edwards 2005 (Continued)
Notes improvements in: UFOV and transfer test for IADL, but not for cogn. factors like Stroop, Trail Making Test, letter
and pattern comparison
Fabre 2002
Methods 4 groups
2 months
pre-post-test
Participants n aerobic=8
n mental=8
n combi=8
n control=8
Interventions aerobic vs. mental training, combination and control
Outcomes physical and cognitive variables Wechsler Memory Scale (memory ratio, paired associated learning, digit span forward,
logiclal memory immediate recall, orientation, general information, mental control, visual reproductions)
Notes control group no changes; cognitive variables improved in 3 training groups, mostly in combined group
Flynn 1990
Methods 2 treatment groups, 1 control
Participants n manual=18
n manual+dicussion=21
Interventions self-studied memory training manual vs. self-studied memory training manual + group discussion
Outcomes - vocabulary subtest of Wechsler Adult Intelligence Scale-Revised

- 5 memory tests: word list, word list travelling, name-face, story recall, verbal digit span forward
Notes
Gratzinger 1990
Methods baseline, after no mnemonic training, after mnemonic training
Participants N=156,
M age=68.42
MMSE>27
Interventions - imagery + mnemonic training

- relaxation + mnemonic
- imagery+judgement+mnemonic
Gratzinger 1990 (Continued)
Outcomes face-name recall

NEO-PI
Notes sign. overall effects but no differences between groups
Hill 1987
Methods 2 weeks group training 8x1h
Participants n training=59
n active controls=17
Interventions mnemonic training vs. development of own learning strategies
Outcomes confidence ratings, name-face recall
Notes
Hill 1988
Methods 6 groups; baseline, after pretraining, after mnemotraining
Participants n imagery+ affective judgement+mnemonic=36
Interventions imagery, relaxation, affective judgement, unspecific training
Outcomes name-faces recall
Notes
Hill 1990
Methods Pre-/Postdesign, 4 groups; 2h+manual+homework
Participants n mem+incentive=16
n memory=14
n active controls+incentive=16
Interventions memory training, active controls, (incentive)
Outcomes study time,

recall time,
word recall
Notes
Hill 1991
Methods 1 day training; 3 groups; baseline, imm. after training, 1h after training, 3 days after training
Participants n story=23
n loci=27
n active controls=21
Interventions narrative story, method of loci and active control condition
Outcomes 26 nouns (free recall)
Notes
Levine 2007
Methods cross-over-design
Participants - Early training group=EG=26

- Late training group=CG=20
Interventions goal management training as cognitive rehabilitation program
Outcomes - simulated real life tasks (SRLT)

- DEX: dysexecutive questionnaire
Notes total score SRLT and subscore available
Mahncke 2006
Methods 1 treatment,
1 treatment control,
1 no contact control
Participants n varies in respect to outcome measures and groups (n=50-56); age range = 60-87
Interventions experimental computer-based training, active computer-based training in auditory language system
Outcomes digit span, global auditory memory score
Notes
Margrett 2006
Methods 6 weeks
individual versus collaborative learning in inductive reasoning
Margrett 2006 (Continued)
Participants n individual training=30

n partner=34
n control (questionnaire)=34
Interventions inductive reasoning
Outcomes letter series test

word series test
letter sets test
Notes
Piccolini 1992
Methods treated vs. nontreated design; 1 month
Participants n treated=12
n non-treated=12
Interventions mnemotechniques: visual tests, internal and external cues
Outcomes verbal and spatial learning, short-term memory, attention, dementia scale, anxiety and depression
Notes stratified, then randomly assigned
Rapp 2002
Methods
Participants meeting criteria for MCI (Petersen et al., 1999) including (1) a self-reported memory complaint, (2) a score on a
standardized memory test at or below the 10th percentile, (3) scores on tests of all other cognitive functions greater
than the 10th percentile, (4) normal global cognitive functioning, (5) no ADL or IADL deficits, and (6) the absence
of dementia. Global cognitive functioning was assessed with the MMSE, perceptions of memory impairment with
the Mermoy Functioning Questionnaire (MFQ), cognitive function with CERAD, perceived control over memory
with the Memory Controllability Inventory and mood was administered with the Profile of Mood States.
n memory training=9
n control=10
Interventions memory training on strategies, info on memory, no contact control group; 6 weeks duration (2 hours/week)
Outcomes immediate and delayed recall of: words, shopping list, name-faces, paragraph
Notes
Rasmusson 1999
Methods 4 groups; pre-/post
Participants n memory group=10

n individual memory=12
n computer=13
n wait list=11
Interventions memory training in groups vs. individualised memory training vs. computer-based individual training vs. wait list
group
Outcomes memory: Hopkins Verbal Learning Test, Rivermead Behavioural Memory Test, Hopkins Prospective Memory Task
questionnaires: Memory Controllability Inventory, Memory Functioning Questionnaire, Geriatric Depression Scale
Notes
Rozzini 2007
Methods One year longitudinal and retrospective comparison study
Participants 59 subjects affected by Mild Cognitive Impairment (MCI) according to Petersen’s criteria including Petersen et al.,
2001), including:
(1) memory complaint, corroborated by an informant;
(2) objective memory impairment;
(3) normal general cognitive functions, as determined by a clinician’s judgement based on a structured interview with
the patients and an informant (Clinical Dementia Rating Scale, CDR score equal to 0.5 with memory box scores of
0.5 or 1) (Hughes et al., 1982) and a Mini Mental State Examination (MMSE) (Folstein et al., 1975) scores greater
than or equal to 24;
(4) no or minimal impairment in activities of daily living (Instrumental Activities of Daily Living, IADL, and Basic
Activities of Daily Living,
BADL) (Lawton and Brody, 1969; Katz et al., 1970) as determined by a clinical interview with the patient and an
informant; and
(5) non cognitive and functional impairment sufficient to meet National Institute of Neurological and Communicative
Disorders and Stroke Alzheimer’s Disease and Related Disorders Association Criteria for AD (McKhann et al., 1984)
, as judged by an experienced AD research clinician.
Depressive mood was excluded by administrating GDS-15 items.
Interventions Fifteen subjects were randomised to receive neuropsychological training plus cholinesterase inhibitors; 22 subjects
cholinesterase inhibitors alone and 22 subjects no treatment; 60 hours over 9 months (1 block = 20 hours/month
with 2 month break); follow-up at 12 months
Outcomes short story recall, letter and semantic verbal fluency, Raven matrices, Rey figure
Notes letter verbal fluency
Schaffer 1992
Methods 2 treatment, 1 no contact control group
Participants n 17/group
Interventions learning skill group (training of attention, organisation, problem solving) vs. social support group (discussions) vs.
control group
Outcomes list recall and recognition, prose recall
Notes
Scogin 1985
Methods high complaint group vs. high complaint control; individual training
Participants n high complaint training=20

n high complaint control=23
Interventions 92-pages manual about memory training
Outcomes immediate and delayed recall of words, shopping list, name-faces; digit span forward, Benton visual retention test
Notes
Scogin 1992
Methods self-taught memory training vs. attention-placebo vs. wait control
Participants n self-taught memory training=22

n attention placebo =23
n wait control=24
Interventions self-taught memory training (117 pages manual)
Outcomes recall of nouns, shopping list, names and faces, paragraph
Notes
Stigsdotter 1989
Methods 2 weeks; pre-/post-design
Participants n multifactor training = 9

n general activation = 9
n control = 10
Stigsdotter 1989 (Continued)
Interventions multifactorial training (loci, imagery, attention, relaxation) versus general cognitive activation, control group
Outcomes - abstract and concrete word recall

- digit span forward
Notes
Stigsdotter 1993 study 1
Methods 8 weeks, group setting
Participants n mulitfactor=10
n unifactor=9
n control=11
Interventions multifactor training vs. unifactor (encoding operations), vs. control no treatment
Outcomes abstract and concrete word recall
Notes
Stigsdotter 1993 study 2
Methods 8 weeks, baseline, post-test, 6 months, 3.5 year follow-up
Participants n multifactor training = 6

n unifactor training = 6
n control = 6
Interventions multifactor training vs. cogn. activation (problem solving, visuospatial skills), versus control no treatment
Outcomes total word recall, long-term retrieval
Notes
Stigsdotter 1995
Methods 2 groups; group setting for 5 weeks
Participants n multifactor=23
n control group=23
Interventions multifactor training vs. control
Outcomes recall of concrete and abstract words, objects and subject-performed tasks
Stigsdotter 1995 (Continued)
Notes
Valentijn 2005
Methods 8 weeks, double baseline, post-test, follow-up
Participants analyzed n:
n group=39
n individual=40
n control=38
Interventions group mem training vs. individual training vs. wait list
Outcomes short story immediate and delayed recall, word recall, total recall score
Notes
Yesavage 1990
Methods 3 goups
Participants n imagery=74
n relax=67
n imagery+judgement
Interventions imagery vs. relaxaton vs. imagery + judgement training
Outcomes word, name-face recall
Notes
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Andrewes 1996 not randomly selected: (“From those who responded to the advertisement, the first 20 women
and 20 men were selected”)
Anschutz 1985 no control group
Anschutz 1987 no control group
Baldelli 1991 no control group
(Continued)
Baltes 1988 not assigned randomly
Baltes 1989 no data (M, SD) available for baseline and post-test
Belleville 2006 not randomly assigned to treatment or control group: “to control for pre-post pracitce effects on repeated
cognitive testing, a new consecutive group (...) was recruited (...).”
Best 1992 no standard deviations available
Bond 2000 stratefied patients into 3 groups with MMSE of 13-17, 18-23, 24-30, but no differentiated data for the
relevant groups available
Boron 2007 assigned according to previous test performances
Brooks 1993 no data on M, SD available
Brooks 1999 age range 55-88
Calero 1997 explicitly low educated sample
Cavallini 2002 not assigned randomly, no control group
Cavallini 2003 no control group
Cerella 2006 no M, SD data available; only t-values
Cipriani 2006 no healthy control groups: MCI vs. AD vs. systsem atrophy patients
De Vreese 1998 age range 50-87; N=59, n=39 with subjective memory complaints, n=20 with objective memory complaints.
Derwinger 2003 matched groups
Derwinger 2005 b matched groups; follow-up study
Dittman-Kohli 1991 no data for cogn. performances at baseline and post-test, only data available for non-cognitive outcome
variables (perceived utility and efficacy of trained tasks)
Erickson 2007 age range 55-80
Fernandez 2005 N=90, but n of the 4 subgroups unclear for baseline, post-test and follow-up. age range 59-87
Finkel 1989 not assigned randomly
Fleischmann 1985 matched groups
Gunther 2003 no control group
(Continued)
Hill 1989 no control group
Israel 1998 no data available on baseline and post-test means and SD
Johnston 1989 not randomly assigned
Kramer 1995 no pre-/post-test data on N, M, SD available
Labouvie-Vief 1976 no baseline data available, only data of immediate and delayed (2 weeks after training) post-test
Lachman 1992 - stratified assignment to 5 treatment conditions.

- no data on subgroups n available
Martin 1998 no control group
McKitrick 1999 - no data M, SD available

- age range: 55-88 years
Mohs 1998 matched groups
Neils-Strunjas 1997 no control group; multiple N interventions design
Olazarán 2004 no differentiated data on MCI
Onyper 2006 not randomly assigned
Oswald 1996 not randomized
Oswald 2006 not randomly assigned
Oswald 2007 not randomly assigned
Ott-Chervet 1998 MMSE>22; not specified; not randomized
Panza 1996 matched control group, no data available
Paxton 2006 assignment unclear? “either or”
Rebok 1989 no control group
Rebok 1996 only mean performances (raw scores) and change in standard scores available; no standard deviations
Rebok 1997 no data available on cognitive measures at pre-/post-test
(Continued)
Saczynski 1992 assigned according to previous performance
Schaie 1987 assigned according to previous performance
Schmidt 1999 fusion of no contact control group and alternative training group
Schmidt 2001 age range 45-84
Schmitt 1981 not randomly assigned
Scogin 1988 no baseline data for treatment group available

3 year follow-up data
Sheikh 1986 no standard deviations at baseline and post-test available

age >55
Sitzer 2006 a review
Small 2006 Mean age of experimental and control group <60, range 35-69
Stine-Morrow 2007 “...randomly assigned participants to either an experimental or control group with the restriction that partners
be assigned together”
Talassi 2007 unclear assignement to treatment and control groups
Valenzuela 2003 no data available on baseline and post-test
Van Gerven 2003 no pre-/post-test design: comparison of multimedia learning, conventional and unimocal learning with young
and old adults
Van Gerven 2006 no pre-/post-test design: complex problem solving training with bimodal and unimodal training methods for
young and old adults
Van Hooren 2007 only short-term post-test and post-test data available, no baseline
Wadley 2006 alternative assignment
Wenisch 2007 no randomized controlled trial
Werner 2000 61.3% of the sample had a diagnosis of dementia. No explicit information on the other subjects available
West 1992 no data on subgroups available
Willis 1986 assigned according to previous performance
(Continued)
Willis 2006 long-term effects (5 y after first training) of cogn. training on ADL; does not meet our inclusion criterias =>
see chapter ’types of outcome measures’
Winningham 2003 assignment unclear “either exposed to intervention or not”
Winocur 2007 psychosocial training. no cognitive training
Wolinksy 2006 b no cognitive outcome measure: study based on ACTIVE and investigates cognitive training and its relation
to health related quality of life 5 years after baseline
Wolinsky 2006 a no cognitive outcome measure: study based on ACTIVE and investigates cognitive training and its relation
to health related quality of life 1 and 2 years after baseline
Wood 1987 matched groups
Yesavage 1983 a no control group
Yesavage 1983 b no M, SD values available; only F-value and significance
Yesavage 1983 c no control group
Yesavage 1984 not randomly assigned
Zarit 1981 study 1+2: no data available on recall performances
DATA AND ANALYSES
Comparison 1. healthy older adults: treatment vs no contact
No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size
1 face-name immediate recall 4 170 Std. Mean Difference (IV, Fixed, 95% CI) 0.12 [-0.19, 0.43]
2 face-name delayed recall 3 119 Std. Mean Difference (IV, Fixed, 95% CI) -0.06 [-0.43, 0.30]
3 visuo-spatial memory 2 59 Std. Mean Difference (IV, Random, 95% CI) 0.58 [-1.01, 2.17]
4 short-term memory 5 370 Std. Mean Difference (IV, Random, 95% CI) 1.10 [-0.41, 2.61]
5 paired associates 3 120 Std. Mean Difference (IV, Random, 95% CI) 0.74 [-0.06, 1.54]
6 immediate recall 11 529 Std. Mean Difference (IV, Random, 95% CI) 0.43 [0.06, 0.81]
7 delayed recall 6 872 Std. Mean Difference (IV, Fixed, 95% CI) 0.39 [0.16, 0.62]
Comparison 2. healthy older adults: treatment versus active control
No. of No. of
1 face-name immediate recall 5 300 Std. Mean Difference (IV, Random, 95% CI) 0.13 [-0.36, 0.61]
2 face-name delayed recall 3 213 Std. Mean Difference (IV, Random, 95% CI) -0.04 [-0.55, 0.47]
3 visuo-spatial 2 133 Std. Mean Difference (IV, Random, 95% CI) -0.42 [-1.26, 0.41]
4 short-term memory 5 426 Std. Mean Difference (IV, Random, 95% CI) 1.09 [-0.70, 2.88]
5 paired associates 4 247 Std. Mean Difference (IV, Fixed, 95% CI) -0.23 [-0.48, 0.02]
6 immediate recall 12 705 Std. Mean Difference (IV, Random, 95% CI) 0.18 [-0.16, 0.52]
7 delayed recall 5 280 Std. Mean Difference (IV, Random, 95% CI) 0.04 [-0.51, 0.58]
Comparison 3. MCI: treatment vs no contact
No. of No. of
1 immediate recall 3 72 Std. Mean Difference (IV, Fixed, 95% CI) 0.50 [0.02, 0.98]
2 delayed recall 2 35 Std. Mean Difference (IV, Fixed, 95% CI) 0.69 [-0.00, 1.39]
3 executive function 1 37 Std. Mean Difference (IV, Fixed, 95% CI) -0.09 [-0.75, 0.57]
Comparison 4. MCI: treatment vs alternative treatment
No. of No. of
1 immediate recall 2 53 Std. Mean Difference (IV, Random, 95% CI) 1.03 [-0.14, 2.19]
2 delayed recall 1 17 Mean Difference (IV, Fixed, 95% CI) 3.40 [-7.52, 14.32]
Analysis 1.1. Comparison 1 healthy older adults: treatment vs no contact, Outcome 1 face-name
immediate recall.
Review: Cognition-based interventions for healthy older people and people with mild cognitive impairment
Comparison: 1 healthy older adults: treatment vs no contact
Outcome: 1 face-name immediate recall
Study or subgroup Favours control no contact control Std. Mean Difference Weight Std. Mean Difference
N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI
Flynn 1990 18 0.89 (3.52) 19 0.89 (3.52) 23.5 % 0.0 [ -0.64, 0.64 ]
Hill 1988 36 3.35 (3.4) 15 1.1 (3.01) 25.6 % 0.67 [ 0.06, 1.29 ]
Scogin 1985 20 2 (3.43) 23 1.94 (4.37) 27.2 % 0.01 [ -0.58, 0.61 ]
Scogin 1992 16 -0.1 (3.8) 23 0.8 (3.56) 23.8 % -0.24 [ -0.88, 0.40 ]
Total (95% CI) 90 80 100.0 % 0.12 [ -0.19, 0.43 ]

Heterogeneity: Chi2 = 4.55, df = 3 (P = 0.21); I2 =34%
Test for overall effect: Z = 0.75 (P = 0.46)
-4 -2 0 2 4
Favours control Favours experimental
Analysis 1.2. Comparison 1 healthy older adults: treatment vs no contact, Outcome 2 face-name delayed
recall.
Outcome: 2 face-name delayed recall
Study or subgroup treatment no contact control Std. Mean Difference Weight Std. Mean Difference
Flynn 1990 18 0.66 (4.9) 19 0.95 (3.66) 31.5 % -0.07 [ -0.71, 0.58 ]
Scogin 1985 20 1.6 (3.74) 23 2 (4.48) 36.4 % -0.09 [ -0.69, 0.51 ]
Scogin 1992 16 0.4 (3.16) 23 0.5 (3.4) 32.1 % -0.03 [ -0.67, 0.61 ]
Total (95% CI) 54 65 100.0 % -0.06 [ -0.43, 0.30 ]

Heterogeneity: Chi2 = 0.02, df = 2 (P = 0.99); I2 =0.0%
-4 -2 0 2 4
Favours experimental Favours control
Analysis 1.3. Comparison 1 healthy older adults: treatment vs no contact, Outcome 3 visuo-spatial memory.
Outcome: 3 visuo-spatial memory
N Mean(SD) N Mean(SD) IV,Random,95% CI IV,Random,95% CI
Fabre 2002 8 2.5 (1.33) 8 0.4 (1.37) 45.4 % 1.47 [ 0.33, 2.61 ]
Scogin 1985 20 0.35 (1.84) 23 0.63 (1.69) 54.6 % -0.16 [ -0.76, 0.44 ]
Total (95% CI) 28 31 100.0 % 0.58 [ -1.01, 2.17 ]

Heterogeneity: Tau2 = 1.11; Chi2 = 6.10, df = 1 (P = 0.01); I2 =84%
-4 -2 0 2 4
favours control Favours experimental
Analysis 1.4. Comparison 1 healthy older adults: treatment vs no contact, Outcome 4 short-term memory.
Outcome: 4 short-term memory
Buiza 2007 85 12.51 (5.22) 85 -1.56 (2.48) 20.4 % 3.43 [ 2.95, 3.90 ]
Fabre 2002 8 -0.1 (1) 8 0 (0.35) 19.1 % -0.13 [ -1.11, 0.86 ]
Flynn 1990 18 0.94 (3.57) 19 0.26 (2.12) 20.0 % 0.23 [ -0.42, 0.88 ]
Mahncke 2006 50 0.7 (0.35) 54 0 (0.3) 20.4 % 2.14 [ 1.65, 2.62 ]
Scogin 1985 20 0.25 (2.47) 23 0.91 (2.282) 20.1 % -0.27 [ -0.88, 0.33 ]
Total (95% CI) 181 189 100.0 % 1.10 [ -0.41, 2.61 ]

Heterogeneity: Tau2 = 2.84; Chi2 = 127.74, df = 4 (P<0.00001); I2 =97%
-10 -5 0 5 10
Analysis 1.5. Comparison 1 healthy older adults: treatment vs no contact, Outcome 5 paired associates.
Outcome: 5 paired associates
Dunlosky 2007 study 2 34 3.9 (6.95) 29 -0.3 (8.85) 40.8 % 0.53 [ 0.02, 1.03 ]
Dunlosky 2007 study1 21 6.1 (15.22) 20 3.6 (13.05) 37.9 % 0.17 [ -0.44, 0.79 ]
Fabre 2002 8 1.8 (0.85) 8 -0.8 (1.37) 21.3 % 2.16 [ 0.85, 3.46 ]
Total (95% CI) 63 57 100.0 % 0.74 [ -0.06, 1.54 ]

-10 -5 0 5 10
Analysis 1.6. Comparison 1 healthy older adults: treatment vs no contact, Outcome 6 immediate recall.
Outcome: 6 immediate recall
Buiza 2007 85 0.28 (6.47) 85 0.57 (6.7) 13.4 % -0.04 [ -0.34, 0.26 ]
Craik 2007 29 0.04 (0.17) 20 -0.09 (0.18) 10.8 % 0.73 [ 0.15, 1.32 ]
Fabre 2002 8 2.6 (1) 8 0.7 (1.5) 6.3 % 1.41 [ 0.28, 2.54 ]
Flynn 1990 18 2.27 (4.81) 19 0.06 (3.33) 10.1 % 0.53 [ -0.13, 1.18 ]
Scogin 1985 20 3.15 (4.52) 23 3.85 (5.14) 10.7 % -0.14 [ -0.74, 0.46 ]
Scogin 1992 16 2.7 (4.4) 23 2 (4.24) 10.3 % 0.16 [ -0.48, 0.80 ]
Stigsdotter 1989 9 9.2 (10.71) 10 -3.3 (11.52) 7.4 % 1.07 [ 0.09, 2.05 ]
Stigsdotter 1993 study 1 6 -12.84 (0.57) 6 -1.33 (0.92) 0.3 % -13.88 [ -20.75, -7.01 ]
Stigsdotter 1993 study 2 10 2.25 (2.34) 11 -0.41 (2.14) 7.7 % 1.14 [ 0.20, 2.08 ]
Stigsdotter 1995 23 5.09 (11.85) 23 0.61 (8.52) 10.8 % 0.43 [ -0.16, 1.01 ]
Valentijn 2005 39 3.87 (8.53) 38 0.82 (8.23) 12.1 % 0.36 [ -0.09, 0.81 ]
Total (95% CI) 263 266 100.0 % 0.43 [ 0.06, 0.81 ]

-4 -2 0 2 4
Analysis 1.7. Comparison 1 healthy older adults: treatment vs no contact, Outcome 7 delayed recall.
Outcome: 7 delayed recall
Craik 2007 29 0.36 (2.43) 20 -0.12 (2.86) 16.5 % 0.18 [ -0.39, 0.75 ]
Flynn 1990 18 2.72 (6.07) 19 1.37 (5.35) 12.9 % 0.23 [ -0.42, 0.88 ]
Scogin 1985 20 4.05 (4.94) 638 0 (4.57) 26.9 % 0.88 [ 0.44, 1.33 ]
Scogin 1992 16 2 (5.4) 23 1.8 (5.14) 13.2 % 0.04 [ -0.60, 0.68 ]
Stigsdotter 1993 study 2 6 19 (14.53) 6 -0.83 (20.98) 3.5 % 1.01 [ -0.22, 2.25 ]
Valentijn 2005 39 0.63 (5.1) 38 -0.47 (6.05) 26.9 % 0.19 [ -0.25, 0.64 ]
Total (95% CI) 128 744 100.0 % 0.39 [ 0.16, 0.62 ]

-10 -5 0 5 10
Analysis 2.1. Comparison 2 healthy older adults: treatment versus active control, Outcome 1 face-name
immediate recall.
Comparison: 2 healthy older adults: treatment versus active control
Outcome: 1 face-name immediate recall
Study or subgroup treatment active control Std. Mean Difference Weight Std. Mean Difference
Flynn 1990 18 0.56 (4.67) 21 2.42 (3.44) 18.8 % -0.45 [ -1.09, 0.19 ]
Gratzinger 1990 50 2 (2.64) 50 2.44 (2.73) 23.6 % -0.16 [ -0.56, 0.23 ]
Hill 1987 59 2.3 (3.11) 17 0.7 (3.21) 20.6 % 0.51 [ -0.04, 1.05 ]
Hill 1988 36 3.4 (3.35) 16 0.2 (3.2) 19.2 % 0.95 [ 0.33, 1.57 ]
Scogin 1992 16 -0.1 (3.8) 17 0.6 (2.82) 17.9 % -0.21 [ -0.89, 0.48 ]
Total (95% CI) 179 121 100.0 % 0.13 [ -0.36, 0.61 ]

-4 -2 0 2 4
Favours active control Favours experimental
Analysis 2.2. Comparison 2 healthy older adults: treatment versus active control, Outcome 2 face-name
delayed recall.
Outcome: 2 face-name delayed recall
Flynn 1990 18 0.66 (4.9) 21 2.76 (3.59) 29.1 % -0.48 [ -1.12, 0.15 ]
Scogin 1992 16 0.4 (3.16) 17 0.8 (2.6) 27.3 % -0.14 [ -0.82, 0.55 ]
Yesavage 1990 74 3.3 (2.73) 67 2.42 (2.76) 43.6 % 0.32 [ -0.01, 0.65 ]
Total (95% CI) 108 105 100.0 % -0.04 [ -0.55, 0.47 ]

-4 -2 0 2 4
Analysis 2.3. Comparison 2 healthy older adults: treatment versus active control, Outcome 3 visuo-spatial.
Outcome: 3 visuo-spatial
Caprio 1996 61 -0.21 (2.4) 56 1.65 (2.58) 63.6 % -0.74 [ -1.12, -0.37 ]
Fabre 2002 8 2.5 (1.33) 8 2.2 (2.47) 36.4 % 0.14 [ -0.84, 1.12 ]
Total (95% CI) 69 64 100.0 % -0.42 [ -1.26, 0.41 ]

-2 -1 0 1 2
Analysis 2.4. Comparison 2 healthy older adults: treatment versus active control, Outcome 4 short-term
memory.
Outcome: 4 short-term memory
Buiza 2007 85 12.51 (5.22) 68 -1.43 (2.1) 20.2 % 3.35 [ 2.86, 3.85 ]
Caprio 1996 61 0.47 (2.01) 56 1.29 (2.15) 20.4 % -0.39 [ -0.76, -0.03 ]
Fabre 2002 8 -0.1 (1) 8 0.7 (0.75) 19.2 % -0.86 [ -1.89, 0.18 ]
Flynn 1990 18 0.94 (3.57) 21 0.19 (3.01) 20.0 % 0.22 [ -0.41, 0.86 ]
Mahncke 2006 50 0.7 (0.35) 51 -0.3 (0.3) 20.1 % 3.05 [ 2.47, 3.63 ]
Total (95% CI) 222 204 100.0 % 1.09 [ -0.70, 2.88 ]

-10 -5 0 5 10
Analysis 2.5. Comparison 2 healthy older adults: treatment versus active control, Outcome 5 paired
associates.
Outcome: 5 paired associates
Caprio 1996 61 0.97 (2.07) 56 1.93 (2.2) 46.8 % -0.45 [ -0.81, -0.08 ]
Dunlosky 2007 study 2 34 3.9 (6.95) 38 3.9 (9.57) 29.5 % 0.0 [ -0.46, 0.46 ]
Dunlosky 2007 study1 21 6.1 (15.22) 21 5.6 (15.82) 17.3 % 0.03 [ -0.57, 0.64 ]
Fabre 2002 8 1.8 (0.85) 8 2.2 (0.79) 6.4 % -0.46 [ -1.46, 0.54 ]
Total (95% CI) 124 123 100.0 % -0.23 [ -0.48, 0.02 ]

-10 -5 0 5 10
Analysis 2.6. Comparison 2 healthy older adults: treatment versus active control, Outcome 6 immediate
recall.
Buiza 2007 85 0.28 (6.47) 68 1.56 (6.81) 11.8 % -0.19 [ -0.51, 0.13 ]
Caprio 1996 61 0.77 (2.3) 56 1.52 (2.24) 11.4 % -0.33 [ -0.69, 0.04 ]
Fabre 2002 8 2.6 (1) 8 1.3 (0.91) 5.5 % 1.29 [ 0.18, 2.39 ]
Flynn 1990 18 2.27 (4.81) 21 4.47 (3.19) 8.9 % -0.54 [ -1.18, 0.11 ]
Hill 1990 14 8.79 (4.65) 16 4.13 (5.57) 8.0 % 0.88 [ 0.12, 1.63 ]
Hill 1991 27 6.18 (5.47) 21 2 (6.19) 9.4 % 0.71 [ 0.12, 1.30 ]
Scogin 1992 16 2.7 (4.4) 17 1.5 (3.14) 8.5 % 0.31 [ -0.38, 1.00 ]
Stigsdotter 1989 9 9.2 (10.71) 9 -0.8 (6.93) 6.1 % 1.06 [ 0.05, 2.06 ]
Stigsdotter 1993 study 1 6 -12.84 (0.57) 6 -1.33 (2.86) 1.4 % -5.15 [ -7.91, -2.39 ]
Stigsdotter 1993 study 2 10 2.25 (2.34) 9 0.61 (2.37) 6.6 % 0.67 [ -0.27, 1.60 ]
Valentijn 2005 39 3.87 (8.53) 40 2.6 (9.59) 10.7 % 0.14 [ -0.30, 0.58 ]
Yesavage 1990 74 2 (4.05) 67 1.99 (4.02) 11.7 % 0.00 [ -0.33, 0.33 ]
Total (95% CI) 367 338 100.0 % 0.18 [ -0.16, 0.52 ]

-10 -5 0 5 10
Analysis 2.7. Comparison 2 healthy older adults: treatment versus active control, Outcome 7 delayed recall.
Caprio 1996 61 0.38 (2.4) 56 1.5 (2.79) 25.3 % -0.43 [ -0.80, -0.06 ]
Flynn 1990 18 2.72 (6.07) 21 6.77 (4.69) 20.3 % -0.74 [ -1.39, -0.09 ]
Scogin 1992 16 2 (5.4) 17 0.5 (3.68) 19.7 % 0.32 [ -0.37, 1.01 ]
Stigsdotter 1993 study 2 6 19 (14.53) 6 2 (5.31) 10.6 % 1.43 [ 0.10, 2.77 ]
Valentijn 2005 39 2 (3) 40 0.95 (3.23) 24.0 % 0.33 [ -0.11, 0.78 ]
Total (95% CI) 140 140 100.0 % 0.04 [ -0.51, 0.58 ]

-20 -10 0 10 20
Analysis 3.1. Comparison 3 MCI: treatment vs no contact, Outcome 1 immediate recall.
Comparison: 3 MCI: treatment vs no contact
Study or subgroup Experimental Control Std. Mean Difference Weight Std. Mean Difference
Rapp 2002 9 3.45 (2.93) 10 2.7 (2.38) 28.0 % 0.27 [ -0.64, 1.18 ]
De Vreese 1996 10 3.3 (9.68) 6 -1.33 (14.53) 21.9 % 0.38 [ -0.65, 1.40 ]
Rozzini 2007 15 3.5 (3.08) 22 1 (3.9) 50.1 % 0.68 [ 0.00, 1.36 ]
Total (95% CI) 34 38 100.0 % 0.50 [ 0.02, 0.98 ]

-10 -5 0 5 10
Analysis 3.2. Comparison 3 MCI: treatment vs no contact, Outcome 2 delayed recall.
Rapp 2002 9 4.88 (3.63) 10 2.8 (2.76) 56.6 % 0.62 [ -0.31, 1.55 ]
De Vreese 1996 10 5.4 (8.83) 6 -3.5 (13.39) 43.4 % 0.79 [ -0.27, 1.85 ]
Total (95% CI) 19 16 100.0 % 0.69 [ 0.00, 1.39 ]

-20 -10 0 10 20
Analysis 3.3. Comparison 3 MCI: treatment vs no contact, Outcome 3 executive function.
Outcome: 3 executive function
Rozzini 2007 15 1.7 (6.81) 22 2.5 (9.94) 100.0 % -0.09 [ -0.75, 0.57 ]
Total (95% CI) 15 22 100.0 % -0.09 [ -0.75, 0.57 ]

Heterogeneity: not applicable
-10 -5 0 5 10
Analysis 4.1. Comparison 4 MCI: treatment vs alternative treatment, Outcome 1 immediate recall.
Comparison: 4 MCI: treatment vs alternative treatment
Study or subgroup experimental alternative treatment Std. Mean Difference Weight Std. Mean Difference
De Vreese 1996 9 13.22 (12.86) 7 7.29 (16.61) 46.1 % 0.38 [ -0.62, 1.38 ]
Rozzini 2007 15 3.5 (3.08) 22 0.3 (0.5) 53.9 % 1.58 [ 0.82, 2.33 ]
Total (95% CI) 24 29 100.0 % 1.03 [ -0.14, 2.19 ]

-20 -10 0 10 20
Analysis 4.2. Comparison 4 MCI: treatment vs alternative treatment, Outcome 2 delayed recall.
Comparison: 4 MCI: treatment vs alternative treatment
Study or subgroup experimental alternative treatment Mean Difference Weight Mean Difference
De Vreese 1996 10 5.4 (8.83) 7 2 (12.75) 100.0 % 3.40 [ -7.52, 14.32 ]
Total (95% CI) 10 7 100.0 % 3.40 [ -7.52, 14.32 ]

Heterogeneity: not applicable
-10 -5 0 5 10
Favours alternative treat Favours experimental
HISTORY
Protocol first published: Issue 4, 2006
Review first published: Issue 1, 2011
Date Event Description
1 September 2008 Amended Converted to new review format.
1 August 2006 Amended August 2006: This protocol replaces the previous protocol “Cognition-based interventions for
people with Mild Cognitive Impairment” (authors Cameron MH, Clare L) and also adds a healthy
population to the review’s scope.
CONTRIBUTIONS OF AUTHORS
MM - all correspondence, drafting of review versions, selection for trials for inclusion/exclusion, extraction of data, entry of data,
interpretation of analyses
LC - drafting of review versions, selection of trials for inclusion/exclusion, interpretation of data analyses
MA, FZ - search for trials, obtaining copies of trial reports, selection of trials for inclusion/exclusion
FZ - extraction of data, entry of data, analysis in RevMan
MC - interpretation of data analyses
Contact editor: Frans Verhey
Consumer editors: Dave Hanbury, Victoria Morgan, Jean Town
DECLARATIONS OF INTEREST
None known.
SOURCES OF SUPPORT
Internal sources
• Institute of Psychology, University of Zurich, Switzerland.
• School of Psychology, University of Wales, Bangor, UK.
External sources
• No sources of support supplied
DIFFERENCES BETWEEN PROTOCOL AND REVIEW

Although this was considered for the analyses, there was not sufficient information to analyse follow-up data, results on transfer effects
of the interventions, and conversion rates of persons with mild cognitive impairment to dementia. We increased our search range to
1970-2007 (originally 1985), and FZ joined the author group.

Memory 2

Uploaded by

Copyright:

Available Formats

Memory 2

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Memory 2

Uploaded by

Copyright:

Available Formats

Cognition-based interventions for healthy older people and

people with mild cognitive impairment (Review)

Martin M, Clare L, Altgassen AM, Cameron MH, Zehnder F

Cognition-based interventions for healthy older people and

Editorial group: Cochrane Dementia and Cognitive Improvement Group.

Data collection and analysis

PLAIN LANGUAGE SUMMARY

To date, most studies investigating the effectiveness of cognitive

See Cochrane Dementia and Cognitive Improvement Group

Cognitive Domain Ability subgroups Availability of pooled data for

healthy older people People with MCI

Memory face-name immediate recall yes not enough data

face-name delayed recall yes not enough data

visuo-spatial memory (Benton, vi- yes not enough data

short term memory (digit span) yes not enough data

paired associates yes not enough data

immediate recall (of words, para- yes yes

delayed recall (of words, para- yes not enough data

prospective memory not enough data not enough data

phonematic verbal fluency not enough data not enough data

abstraction not enough data not enough data

inductive reasoning not enough data not enough data

Attention - not enough data not enough data

Speed - not enough data not enough data

Table 1: Overview of evaluated cognitive domains and ability

Methods single-blind design; no contact control group;training lasts 5-6 weeks

Participants - n no contact control=704

Interventions problem-solving training, memory strategies, speed training

Outcomes problem solving, verbal episodic memory, speed;

Participants - n exp structured 1=85

Interventions attention, orientation, memory, language, visuoconstruction, visuomanual coordination, praxis

Notes data available: initial, annual, biannual

Methods memory training versus alternative training; no no contact control group

Participants age range 65-76 years

Interventions cognitive restructuration technique versus traditional memory training

Participants n early =29

Interventions memory skills training

Outcomes alpha span (working memory) total score,

Interventions drug treatment and memory training

Outcomes Randt Memory Test:acquisition, delayed recall, memory index

Methods 2 experimental groups and 1 no contact control; 5 weeks, group setting

Outcomes free recall of digit spans

Dunlosky 2007 study 2

Participants n strategy/imagery training=34

Interventions strategy/imagery vs. self-monitoring training and no contact control

Dunlosky 2007 study1

Methods 2 sessions - length?

Interventions self-monitoring approach for improving older adult learning

Interventions speed of processing training

Outcomes cognitive performance and IADL: UFOV,

Interventions aerobic vs. mental training, combination and control

Methods 2 treatment groups, 1 control

Outcomes - vocabulary subtest of Wechsler Adult Intelligence Scale-Revised

Methods baseline, after no mnemonic training, after mnemonic training

Interventions - imagery + mnemonic training

Outcomes face-name recall

Notes sign. overall effects but no differences between groups

Methods 2 weeks group training 8x1h

Interventions mnemonic training vs. development of own learning strategies

Outcomes confidence ratings, name-face recall