Table 1.

Study and participant characteristics of the included studies

Study	Sample				Training		Outcomes	Prognostic factors
Study design Initial sample size for the experimental group Dropout and reasons	Age (years, M, SD)	Sex		Education(years, M, SD)	Description of memory training—content and frequency	Total length of training in minutes	Definition and method of assessment Timing of outcome assessment	Definition and methods
		♂	♀
Pesce et al. [20] Stratified randomized study n = 30 n = 29	70.40 (7.00)	14	15	9.60 (1.80)	Method of loci and general strategies. 24 weeks, 2 times a week for 1 h	2880	RAVLT, MMSE	Antioxidant levels assessed with the Biological Antioxidant potential Test; reactive oxygen metabolites derivative compounds assessed with the d-ROMs Test
O'Hara et al. [9] Non-randomized, non-controlled longitudinal study n = 531 n = 419 due to several reasons at 5-year follow-up	73.73(7.62)	34	78	15.56 (2.79)	Method of loci. 2 weeks, 5 times a week for 2 h	1200	Number of words correctly recalled, number of words correctly recalled in order. Assessed at baseline and 5-year follow-up measurement	Pre-training, gain scores following training, age, education, reported use of mnemonic at follow-up, type of pre-training (standard vs. comprehensive) and length of training.
Mohs et al. [21] RCT n = 68 n = n.a.	78.30 (7.40)	15	53	16.00 (2.70)	Structured memory training focusing on memory improvement and different strategies. Nine 90-min sessions	810	Verbal memory assessed with CVLT, non-verbal memory assessed with BFLT. Assessed 2 times at baseline, at post-test, 3 months and 6 months follow-up	Age, education, gender, subjective reported memory assessed with the MFI and the MFQ
Kirchhoff et al. [22] Non-randomized, non-controlled longitudinal study n = 16 n = 2 due to technical difficulties	72.00 (66–81)	7	7	14.70 (2.90)	Memory strategy training and practice. 2 training sessions	Missing information	Memory retrieval using Remember/Know/New recognition memory decisions Assessed at pre-training and post-training	Hippocampal activity
Kirchhoff et al. [23] Controlled trial n = 16 n = n.a.	71.9 (66–81)	8	8	14.8 (2.7)	Memory strategy training and practice 2 training sessions.	Missing information	Recognition memory using Remember/Know/New recognition memory decisions. Assessed at pre-training and post-training	Activity in prefrontal cortex, left lateral temporal cortex.
Leahy et al. [24, 25] RCT n = 22 n = 1	74.77 (6.57)	8	13	18.77 (2.62)	Memory specificity training to improve the specificity of older adults’ retrieval of autobiographical memories by providing systematic practice. 4 weeks, once a week for 60 min	240	Autobiographical memory specificity. Assessed at pre-test, post-test, and 3 months follow-up.	Memory specificity assessed with MEPS, functional limitations assessed with FLP, self-rated depression assessed with HADS, independence assessed with IADL
Andrewes et al. [26] RCT controlled for sex n = 20 n = 3	60–70 years	10	10	Some secondary schooling: n = 3 Secondary school + trade qualifications: n = 5 Complete secondary school: n = 6 Began tertiary school: n = 6	Memory handbook training for face-name and prospective memory areas; independently implemented at home 4 weeks, 30 min per session	Missing information	Improvement in: Face-name Test, Laboratory Prospective Memory Assessment, Everyday Prospective Memory Assessment. Assessed at pre-test, post-test and 4-month FU	RAVLT, Warrington Forced-Choice Recognition for Faces, BDI, NART, Mattis Dementia Rating Scale.
Anschutz et al. [27] Non-randomized, non-controlled longitudinal study n = 10 n = 1 due to severe illness	73.50 (n.a.)	2	7	10.70 (n.a.)	Method of loci No information on training duration and frequency	Missing information	Free-recall of two lists and recognition of two lists consisting of 12 nouns each. Assessed at pre-test and 34 months after finishing the training	Free-recall pre-test, free recall list 1, age
Bissig and Lustig [28] Non-randomized, non-controlled longitudinal study n = 19 n = 1 due to low accuracy of studied words	74.50 (6.10)	n.a.	n.a.	18.00 (3.30)	Modified recollection training procedure 2 weeks, 4 sessions per day at 7 days	Missing information	Ranking: participants were ranked by final lag level (lag between lure repetitions). Assessed and adapted individually during each training performance	Age, crystallized intelligence
Bråthen et al. [29] Controlled trial n = 126 n = 3	Old: 73.40 (3.00)	Old: 29	Old: 49	Old: 14.70 (2.90)	Learning and practicing the Method of loci technique aiming to improve episodic memory performance 10 weeks, once a week + 8 weekly online home assignments	Missing information	Memory improvement: change in correct written recall of word list consisting of 100 nouns. Assessed at pre-test and post-test	Cortical volume, hippocampal volume, ALFF, fALFF
Brooks et al. [8] RCT n = 224 Dropout not reported	68.58 (7.05)	n.a.	n.a.	15.33 (2.58)	Pre-training: imagery training, verbal elaboration and relaxation. Name-Face Mnemonic: three-step mnemonic Method of loci: method of loci for serial word recall. 2 weeks, 5 times a week for 120 min)	1200	Proper name recall task, word recall task (16 common words). Assessed at pre-test and post-test	Pretraining, pretest score, age, length of training, pretraining x length
Clark et al. [30] Multi-site RCT (ACTIVE) n = n.a. n = n.a.	No demographics separately for the memory training groups were reported.				Memory training focused on improving verbal episodic memory through instruction and practice in strategy use 6 weeks, 10 60-min sessions	600	HVLT, RAVL, RBMT. Measured at baseline, immediate post-training, 1-;2-, 3-; 5-, and 10-year FU	Obesity, determined from BMI (in kg/m2) computed from measured height and weight data obtained at baseline
Clark et al. [31] Multi-site RCT (ACTIVE) n = n.a. n = n.a.	No demographics separately for the memory training groups were reported.				Memory training focused on improving verbal episodic memory through instruction and practice in strategy use 6 weeks, 10 60-min sessions	600	Hopkins Verbal Learning Test, RAVL, RBMT. Measured at baseline, immediate post-training, 1-;2-, 3-; 5-, and 10-year FU	Education (self-reported as years of completed schooling)
de Lange et al., [32] Controlled trial n = 76 n = 9 due to time constraints	73.60 (3.00)	25	51	15.00 (2.70)	Learning and practicing the Method of Loci technique aiming to improve episodic memory performance. 10 weeks, once a week + 8 weekly online home assignments	Missing information	Memory improvement: Word list recall	Interindividual variability in white matter microstructure
de Lange et al., [33] Controlled trial n = 44 n = 0	73.30 (2.70)	21	23	15.70 (3.10)	Learning and practicing the Method of Loci technique aiming to improve episodic memory performance. 10 weeks, once a week + 8 weekly online home assignments	Missing information	Memory improvement: word list test (100 words)	White matter microstructure
Tomaszewski Farias et al. [34] Multi-site RCT (ACTIVE) n = n.a. n = n.a.	No demographics separately for the memory training groups were reported				Memory training focused on improving verbal episodic memory through instruction and practice in strategy use 6 weeks, 10 60-min sessions	600	Memory factor: Immediate recall HVLT, RAVLT, paragraph recall, RBMT	Instrumental activities of daily living, 18 questions of the Minimum Dataset Home Care scale
Finkel and Yesavage [35] Controlled trial n = 77 n = 16 due to illness (n = 5), frustration (n = 7), bad weather (n = 2), no reason (n = 1)	71.29 (6.31)	30%	70%	n.a.	Method of loci No information on training duration and frequency	Missing information	Memory improvement gain scores of a list of 16 common words recall	Age, education, MMSE score, depression score, neuroticism and extraversion scale of the NEO-PI
Hampstead et al. [36] RCT n = 12 n = 1 due to ongoing disease	73.20 (7.70)	n.a.	n.a.	16.10 (3.40)	Object Location Assignment encoding and retrieval with mnemonic strategy from a cognitive rehabilitation program 2 weeks, 5 sessions + 1 follow-up session one month later	Missing information	Modified change score of Object Location Assignment accuracy	Medial temporal lobe volumetrics (hippocampus, amygdala, inferior lateral ventricles), standardized neuropsychological measures (RBANS Delayed Memory Index, TMT B)
Hill et al. [37] Controlled trial n = 59 n = n.a.	67.80 (7.50)	n.a.	n.a.	5.80 (1.10)	Mnemonic training 2 weeks, twice a week for 120 min	1680	Recall performance in name-face recall	Rated confidence (perceived confidence in recalling the names of unfamiliar faces).
Hill et al. [38] Non-randomized, non-controlled longitudinal study n = 102 n = n.a.	75.40 (10.50)	32	70	n.a.	Name- and face and list-learning program using an imagery and judgment technique and method of loci method. 2 weeks, 7 times a week for 120 min	1680	Improvement in Name-Face recall, Improvement in List-Recall	MMSE.
Leahy, Ridout, and Holland, [24] RCT n = 20 n = 1 due to unrelated health problems	76.85 (5.27)	6	14	17.75 (2.65)	Memory flexibility program 4 weeks, once a week for 60 min	240	Autobiographical memory specificity in the AMT. Assessed at pre-test, post-test, and 3 month FU.	Baseline cognitive flexibility measured with the verbal fluency sub-score of ACE-III.
López-Higes et al. [39] RCT n = 50 n = 0	ApoE 4 carriers: 71.64 (5.72) Non-carriers: 71.68 (5.65)	n.a.	n.a.	n.a.	Memory training consisting of cognitive stimulation, memory concepts, management of forgetting everyday experiences, meta-memory training 3 months, 30 90-min sessions	2700	Logical Memory and Word List from WMS-III	Apolipoprotein E genotyping
McDougall et al., [40] RCT n = 135 Loss to post-test: n = 8 Loss to FU: n = 12 Loss to end of study: n = 8	74.69 (5.74)	30	105	13.39 (3.90)	CBMEM-based intervention, based on the four components of self-efficacy theory 4 weeks, twice a week including 8 sessions and 4 booster sessions	720	HVLT-R, BVMT-R, RBMT. All outcome measures were administered at baseline, post-class (2 months after baseline), post-booster (6 months), post-classroom FU (14) and at the end of study (24 months)	Ethnicity, group assignment, time, education
McDougall et al., [41] RCT n = 135 Loss to post-test: n = 8 Loss to FU: n = 12 Loss to end of study: n = 8	74.69 (5.74)	30	105	13.39 (3.90)	CBMEM-based intervention, based on the four components of self-efficacy theory 4 weeks, twice a week including 8 sessions and 4 booster sessions	720	Relative gains in HVLT-R, RBMT All outcome measures were administered at baseline, post-class (2 months after baseline), post-booster (6 months), post-classroom FU (14) and at the end of study (24 months)	Age, education, racial/ethnic group
Neely & Bäckman [42] RCT n = 23 n = n.a.	73.00 (4.20)	4	19	9.90 (3.10)	Encoding operations including interactive imagery and method of loci; attention training, relaxation training. Training was conducted in groups with 11–12 subjects, met twice a week for 5 consecutive weeks, each session lasted 1.5 h	900	Recall of concrete words, recall of objects, recall of subject-performed tasks, recall of abstract words Assessed at pre-test, post-test directly after training, 6 months FU	Pretest score for each dependent variable, MMSE score, age, years of education
O’Hara et al., [43] Non-randomized, non-controlled longitudinal study n = 212 n = 113	74.00 (7.90)	68	32	15.50 (2.70)	Memory training was not further described. Missing information on duration and frequency.	Missing information	BVRT, Logical Memory Test, Associate Learning Test, List-learning test. Assessed at baseline and FU 4-5 years after memory training.	Apolipoprotein E genotyping.
Park et al. [7] RCT n = 39 n = n.a.	69.81 (4.90)	11	28	11.41 (4.31)	Multi-strategic memory training. 10 sessions once a week, each session lasted 1.5 h	900	Elderly verbal learning test of the EMS to assess verbal memory; Simple Rey Figure Test of the EMS to assess non-verbal memory. Assessed at pre-test and post-test (within 3 months after finishing the training)	All baseline values of the scores of neuropsychological tests, age, gender, years of education
Rosi et al., [44] Non-randomized, non-controlled longitudinal study n = 44 n = n.a.	68.73 (6.05)	n.a.	n.a.	11.36 (3.50)	Memory training program. 6 weeks, once a week for 60 minutes.	360	Word list learning (memory practiced task), grocery list learning (memory non-practiced task), associative learning Assessed at pre-test and post-test.	Vocabulary test, Raven standard progressive matrices, listening span test, letter comparison, age
Sandberg et al. [45] Non-randomized, non-controlled longitudinal study n = 112 n = 18 due to various reasons	70.90 (6.70)	38	56	11.90 (3.70)	Mnemonic training was based on the Swedish version of the number-consonant mnemonic task 5 times, twice a week	600	Number recall. Assessed at pre-test, post-test and FU	Three measures of episodic memory (free recall of concrete nouns, free recall of abstract nouns, paired-associate recall), three measures of working memory (listening span, two versions of computation span), nine measures of processing speed, two measures of verbal knowledge, depression (ZSRDS), vocabulary

All reported values regarding sample size, dropouts, and sociodemographic variables only refer to the memory training groups. For the variables age (in years) and education (in years) means and standard deviations were displayed, when reported. Otherwise, ranges and/or absolute numbers are stated

RAVLT Rey Auditory Verbal Learning Task, BDI Beck Depression Inventory, NART National Adult Reading Test, dROMs reactive oxygen metabolites derivative compounds, FU follow-up, ALFF amplitude of low-frequency fluctuation, fALFF fractional amplitude of low-frequency fluctuation, BMI body mass index, MMSE Mini-Mental State Examination, NEO-PI NEO Personality Inventory, RBANS Repeatable Battery for the Assessment of Neuropsychological Status, TMT B Trial Making Test Version B, AMT Autobiographical Memory Task, ACE-III Addenbrooke’s Cognitive Examination-III, CBMEM Cognitive Behavioral Model of Everyday Memory, HVLT-R Hopkins Verbal Learning Test-Revised, RBMT Rivermead Behavioral Memory Test, BVMT-R Brief Visuospatial Memory Test revised, EMS Elderly Memory Disorder Scale, BVRT Revised Benton Visual Retention Test, WMS-III Wechsler Memory Scale III, HVLT Hopkins Verbal learning task, MEPS means end problem solving procedure, FLP functional limitation profile, HADS Hospital Anxiety and Depression Scale, IADL instrumental and basic activities of daily living, CVLT California Verbal Learning Test, BFLT Biber Figure Learning Test, MFI memory controllability inventory, MFQ Memory Functioning Questionnaire, ZSRDS Zung Self-Rating Depression Scale