Cognitive decline is one of the most feared aspects of aging (Deary et al, 2009). Cognitive deficits may result from diseases of the central nervous system or from age-related neuronal changes (Reuter-Lorenz and Park, 2014). Age-related changes can decrease older people’s ability to live autonomously as well as their overall quality of life (Blazer et al, 2015). Thus, maintaining one’s cognitive abilities is key to preserving one’s quality of life as we age.
One possible way to preserve or even improve cognition is to participate in a regimen of cognitive training that involves practicing core cognitive abilities (Simons et al, 2016). Cognitive training requires repeated practice and taps into specific cognitive functions (Bamidis et al, 2014). Recent meta-analyses and reviews have shown that training can be effective in improving cognitive functions of healthy older adults (eg, Zhu et al, 2016). It may even reduce neuropsychological symptoms in patients with mild cognitive impairment (MCI) and dementia (eg, Lampit et al, 2014; Reijnders et al, 2013).
However, all cognitive training programs do not have equal success. Researchers have proposed several reasons why some programs are more effective than others. One explanation is related to the specific characteristics of the training program, such as its length, its intensity, and whether it involves individual- or group-based training (Simons et al, 2016). Another possible reason why some programs fail depends on whether the training is scientifically based and well structured. According to Bahar-Fuchs et al (2013), a structured program follows a scientifically based manual of instruction, in which the structure and content of each training session is described in detail. A structured program also includes individual tasks, group tasks, and group games focusing on specific cognitive functions. The interventions used in the ACTIVE trial, the largest brain-training intervention study with healthy older adults to date (Ball et al, 2013; Willis and Caskie, 2013), are an example of structured training.
Unstructured training is often referred to as “unspecific brain jogging.” In this type of training program, cognitive tasks are randomly combined, without special focus on a specific cognitive domain or target group (Simons et al, 2016). Brain jogging is thus mainly directed at practicing core cognitive abilities, and its goal is to improve performance on more than the usually trained key cognitive tasks (ie, memory, attention, executive functions), including performance of everyday activities (Simons et al, 2016). Notably, only a few studies have referred to their cognitive training type as structured or unstructured, but a program can be categorized according to the description provided here.
Previous studies have examined the effectiveness of cognitive training in relation to specific program characteristics such as training dosage (Bamidis et al, 2015) or structure. While some studies have found that the best effects were achieved with a structured training program in both healthy adults (Jaeggi et al, 2008) and patients with Parkinson disease (Petrelli et al, 2014), these findings are preliminary and need further research. Recent studies have shown that multidomain programs seem to be more promising than training that focuses on only a single domain for inducing lasting cognitive benefits in healthy older adults (Cheng et al, 2012; Rahe et al, 2015b).
One further important aspect that seems underinvestigated is which individuals benefit from or respond best to cognitive training (Fairchild et al, 2013). This information is important because the identification of factors found to improve and maintain training effects could inform which type of training program is best suited for an individual (Langbaum et al, 2009). Several variables may have predictive value for cognitive improvement in healthy older adults resulting from cognitive training. For example, studies have reported that individuals with a low baseline neuropsychological test performance benefitted more from training compared with individuals with a high baseline neuropsychological test performance (Langbaum et al, 2009; Rahe et al, 2015a; Whitlock et al, 2012; Zinke et al, 2014). Another study found the opposite: Individuals with a high baseline neuropsychological test performance benefitted most from cognitive training (Fairchild et al, 2013). Furthermore, participants with higher self-rated health (Rebok et al, 2013), younger age (Dorfman and Ager, 1989; Zinke et al, 2014), and higher levels of education (Langbaum et al, 2009) have been found to benefit more from cognitive training. Booster sessions, which are training sessions that participants receive after the official training is completed to maintain and refresh practices and exercises learned during the program, have also provided positive training effects (Ball et al, 2013; Willis et al, 2006). Altogether, the data are far from conclusive, and more research is needed.
Another factor possibly influencing the outcomes of cognitive training, especially for memory, is the apolipoprotein E4 (apoE4) allele, which is a well-known risk factor for Alzheimer disease (Liu et al, 2013). A recent study showed that carrying the apoE4 allele could be used to predict outcomes in a cognitive stimulation program in patients with MCI, finding that carriers of the apoE4 allele showed less improvement in memory than noncarriers (Binetti et al, 2013). Furthermore Rahe et al (2015a), showed that not carrying the apoE4 allele predicted training success in a combined cognitive-physical training intervention in healthy older adults in an executive task. Notably, a meta-analysis revealed that healthy older adults who are carriers of the apoE4 allele performed significantly worse on measures of episodic memory and overall global cognitive ability than noncarriers (Wisdom et al, 2011). The question remains whether carrying the apoE4 allele also reduces cognitive plasticity induced by cognitive training in healthy older adults.
In view of the current state of knowledge of cognitive training, the aim of our study was to compare the effects of a structured cognitive training program that was specifically developed for elderly people to include tasks for enhancing the age-sensitive cognitive domains of verbal memory, attention, and executive functions with the effects of an unstructured brain jogging program on the same outcome measures. Results were also compared to a wait list control group. We hypothesized that the structured cognitive training program would lead to improvements in the tested cognitive domains compared with the active (unstructured training) group and a passive (wait list) control group (hypothesis 1). Another aim was to identify possible predictors for training success. We hypothesized that age, sex, education, carrying the apoE4 allele, and baseline cognitive performance on neuropsychological tests would be predictors for training success (hypothesis 2).
We recruited study participants via brochures and notices in local clinics, churches, and drug stores as well as personal contacts with relatives of patients who had been treated at the University Hospital of Cologne. The participants were randomly divided (using the online Research Randomizer, http://www.randomizer.org) into two groups: one that would receive structured training using the NEUROvitalis training program (Baller et al, 2009) and one that would receive unstructured training using an inhouse-developed training program called Mentally Fit. To allow us to look for test-retest effects, we included a control group, which was nonrandomized as a result of planning difficulties in the clinical daily routine. A total of 105 healthy older adults (NEUROvitalis group: n=35; Mentally Fit group: n=35; control group: n=35) were analyzed to evaluate whether cognitive training had an effect on cognitive domains.
To be included in the study, participants had to be healthy older adults, aged 50 to 85 years, who were native German speakers. Exclusion criteria were suspected dementia (DemTect ≤9 [Kessler et al, 2000]), suspected depression (Becks Depression Inventory II ≥ 14 [Beck et al, 1961]), and/or other neurologic or psychiatric diseases, as well as insufficiently corrected hearing or sight. Hearing, vision, and neurologic and psychiatric diseases were screened via participants’ self-reports. The participants in the control group were matched for age, sex, and education to the participants in the other two groups.
Separately, to create a larger database for the analysis of predictors, we recruited an additional 45 healthy older adults who had previously received NEUROvitalis training but were not part of our analyses of training effects. In other words, the total sample size for the predictor analysis was 80: 35 from our randomized controlled trial’s NEUROvitalis group plus 45 additional NEUROvitalis participants. Of these 80 participants, information on apoE4 carrier status was obtained from the 35 participants who were part of the randomized study: apoE4 carriers=9 and apoE4 noncarriers=29.
Trial Design and Study Setting
Using a randomized controlled study design (following CONSORT guidelines; Schulz et al, 2010), we compared the effects of the NEUROvitalis training program on the domains of verbal memory (word list learning with short-term and long-term recall), attention, and executive functions (set-shifting) with the effects of the Mentally Fit brain jogging program in healthy older adults serving as an active control group. To control for retest effects, a passive control group matched according to age, sex, and education was included in the analysis but assessed separately, making this part of the study a controlled study design. All data were assessed at the Memory Clinic of the University Hospital of Cologne and the University of Vechta from 2007 to 2010.
To enlarge the database for the analysis of predictors, all of the participants who were trained with the NEUROvitalis program at the Memory Clinic of the University Hospital of Cologne from 2007 to 2012 were included for that part of the study. The same inclusion and exclusion criteria for all participants were used. The study was conducted in compliance with the Helsinki Declaration and was approved by the local ethics committee of the University of Cologne. Written informed consent was obtained from each participant. Participants in the two training groups were blinded for the treatment so that they did not know whether they received the structured or unstructured training; participants in the control group were not blinded. The outcome raters of the neuropsychological tests were also blinded to the person’s group assignment; the NEUROvitalis and Mentally Fit trainers who conducted the treatment were not.
Both training programs had the same duration and frequency of training sessions (ie, a total of 12 sessions, each lasting 90 minutes; training was conducted on two different days a week for 6 weeks). Participants were trained in small groups of three to eight people. For both programs, experienced group leaders took responsibility for organizing and carrying out the training. Group tasks and games always involved all of the participants, and individual tasks were supervised by the group leader. Both programs have been described in detail in previous reports (Petrelli et al, 2014; Rahe et al, 2015a), but we provide a general description here.
The multidomain NEUROvitalis cognitive training program has shown promising results in previous studies, in both patients with Parkinson disease and healthy older adults (Petrelli et al, 2014; Rahe et al, 2015a, 2015b). With the NEUROvitalis program, an experienced psychologist conducts training that focuses on specific cognitive functions (eg, memory, attention, and executive functions). Each NEUROvitalis session is described in a detailed and structured manual (Baller et al, 2009; Petrelli et al, 2014). The training consists of different components, including a psychoeducational theoretical part, group and individual tasks, and activation games (ie, games that are designed to target a specific cognitive domain). All of the sessions are similarly arranged: They start with an address of welcome and feedback from the last session; next, the trainer discusses the last homework (cognitive tasks that the participants do at home) with the participants and introduces the topic for the present session (eg, What is memory?). In the psychoeducational part of the NEUROvitalis program, the participants learn about different topics such as mental capacity, attention, memory, working memory, memory strategies, memory for names and faces, appointments and transactions, remembering texts, planning, and risk and protective factors of cognitive aging. Next, the group works on different tasks and plays activation games, which typically fit with the previously discussed topic. For example, the activation game “Category-Memory” trains the domains of memory and categorical thinking. Category-Memory is similar to a classical memory game, but instead of two identical cards, it contains picture cards and (word) category cards (eg, one card shows a picture of a car; the matching card has the name car on it). Each picture card has to be paired with one of four categories (ie, fruits, animals, occupations, vehicles).
The NEUROvitalis training materials have two levels of difficulty so that the training can be adjusted according to the participants’ abilities. Healthy older adults usually train with the more difficult Level 2 exercises, as was the case throughout the training in the present study.
In the Mentally Fit program, cognitive domains are not addressed in focused sessions; rather, individual and group tasks that train memory, attention, and more abstract cognitive functions are combined randomly over the course of the entire program. Instead of the psychoeducation that is part of the NEUROvitalis program, Mentally Fit sessions contain group conversations about topics that are proposed by the trainer or the participants themselves. The cognitive tasks that make up the Mentally Fit training program were selected from a representative choice of tasks of seven frequently used German brain trainings (Baller, 2003; Boos et al, 2007; Halbach, 1998; Normann, 1996; Oppolzer, 1998; Stengel, 1997; Tanklage, 2009) in addition to one group game (Barbian et al, 1995). Two independent experts rated each task of every program with regard to the cognitive domain being trained. The frequency of the tasks that trained each domain was then calculated. Based on these results, Mentally Fit was composed of a representative choice of tasks, which were then randomly put together (Petrelli et al, 2014).
Table 1 presents a detailed overview of the two training programs. The two programs are similar in their frequency and intensity of training sessions, and they train the same cognitive domains—memory, attention, and executive functions. However, the two differ in that the Mentally Fit program trains an even broader spectrum of domains than the NEUROvitalis program. The two programs also differ in their structure and in the type of exercises that are conducted: NEUROvitalis follows a clear structure, with specific domains being trained in single sessions; Mentally Fit is randomly arranged, with random exercises. The two programs also differ in the additional part consisting of psychoeducation with predefined scientifically based topics in the NEUROvitalis training program versus conversations on different topics chosen by the participants or the trainer in the Mentally Fit training program.
We used a battery of neuropsychological tests to assess each participant’s cognitive functioning. The primary outcome domains were memory (verbal short term and verbal long term), attention, and set-shifting as an executive function. The scores of the DemTect subtests Word List, Immediate Recall and Word List, Delayed Recall (Kessler et al, 2000), were used to measure verbal short-term and long-term memory, respectively. The total score of the Brief Test of Attention (Schretlen et al, 1996) was used to measure attention. To measure set-shifting, we calculated the difference between the Trail Making Test part A and part B scores (Reitan, 1958). To minimize retest effects, a parallel version of the DemTect was used at the posttest.
Participants were tested within 10 days before (pretest) and 10 days after (posttest) the training program. In the control group, pre- and posttests were conducted in a parallel time interval without training. The neuropsychological assessments were conducted in a standardized face-to-face situation. The tests were administered by experienced psychologists who were trained in test application and scoring.
For our statistical analysis, we used SPSS 24 for Windows. Normal distributions were tested using the Kolmogorov-Smirnov test. Baseline characteristics (demographics, overall cognitive state) were calculated for each intervention group and were compared between groups. We used one-way ANOVAs to compare age, DemTect total score (as a measure for general cognitive state), and education, and chi-square tests to compare sex distribution and apoE4 state, each with a significance level of α=0.05. G*Power (http://www.gpower.hhu.de) was used to estimate the achieved power with a post hoc analysis (Erdfelder et al, 1996).
Changes from pre- to posttest were analyzed using 2×3 ANOVAs for repeated measures for the domains of verbal memory (short term and long term), attention, and set-shifting as an executive function (hypothesis 1). The within-subject variable, time, had two levels: pre- and posttest. The between-subjects variable, training, had three levels: NEUROvitalis, Mentally Fit, and control group. The effect size, partial η² (η²p), was reported, indicating a small (η²p>0.01), moderate (η²p>0.06), or strong effect (η²p>0.14), as defined by Field (2009).
To analyze the predictors of cognitive improvement from training (hypothesis 2), we used backward multiple regressions to ensure achievement of best model fit considering each relevant predictor. We conducted regression analyses for the NEUROvitalis training group only (n=80: 35 from our randomized controlled trial’s NEUROvitalis group plus an additional 45 NEUROvitalis-trained participants). Changes in scores (posttest–pretest) of the cognitive variables—verbal short-term memory, verbal long-term memory, attention, and executive functions—were calculated and were used as dependent variables. The predictors age, sex, education, and baseline level of performance in the investigated domain were integrated in all four of the conducted regression models. (For further details on the results of the regression analyses, see Supplemental Digital Content 1, https://links.lww.com/CBN/A75.)
Because of the small number of participants for whom apoE4 samples were available, apoE4 carrier status could not be integrated in the regression models. Instead, for apoE4 subgroup analysis, we calculated t tests to compare the change scores (posttest–pretest) between carriers and noncarriers of apoE4 for all outcomes. Figure 1 provides an overview of the flow of participants throughout the study.
The study began with a total of 105 healthy older adults. Six of these were excluded from analysis (NEUROvitalis: n=0; Mentally Fit: n=4; control group: n=2) because they participated in less than 80% of the training sessions due to time constraints, leaving 99 participants in the final analysis. Table 2 shows the baseline characteristics of the three study groups. The three groups did not differ at baseline in age (F2,99=3.03, P=0.053), sex (χ2=3.77, P=0.291), education (F2,99=0.61, P=0.547), or cognitive status (F2,99=2.273, P=0.108). Table 3 provides an overview of the means and SDs of the cognitive outcomes in all of the groups at pre- and posttest.
Structured Versus Unstructured Cognitive Training (Active Control) Versus Passive Control Group
Significant interaction effects of Time×Training were found in favor of the NEUROvitalis group for verbal short-term memory, F2,96=7.59, P=0.001, η²p=0.14, but not for verbal long-term memory, F2,96=1.34, P=0.266, η²p=0.03; attention, F2,87=1.91, P=0.154, η²p=0.04; or executive functions, F2,95=1.27, P=0.284, η²p=0.03. Overall analyses revealed significant within-subject effects of time for verbal short-term memory, F1,66=10.04, P=0.002, η²p=0.10, but no significant within-subject effects of time could be found for verbal long-term memory, F1,96=1.49, P=0.225, η²p=0.02; attention, F1,87=1.34, P=.542, η²p=0.00; or executive functions, F1,95=0.26, P=0.614, η²p=0.00. Overall analyses also revealed significant between-subject effects of training for verbal short-term memory, F2,96=8.65, P<0.001, η²p=0.15, and attention, F2,87=5.39, P<0.05, η²p=0.11, but there were no significant between-subject effects of training for verbal long-term memory, F2,96=2.12, P=0.126, η²p=0.04, or executive functions, F2,95=0.38, P=0.685, η²p=0.01. All significant within- and between-subject effects were in favor of the NEUROvitalis training program. No other significant effects were found. On a descriptive level, a trend for improvement in attention was noticed in favor of the NEUROvitalis training group compared with the Mentally Fit training group and the control group.
Analysis of Predictors
Because the best gains were clearly realized by the study group who used the structured cognitive training, NEUROvitalis, we performed an exploratory backward regression analysis of predictors using only participants with NEUROvitalis training (n=80). The main results of this analysis were as follows:
- Low baseline performance was a predictor for gains in verbal short-term memory (B=–0.58), verbal long-term memory (B=–0.50), attention (B=–0.68), and executive functions/set-shifting (B=–0.46).
- Being a woman was a predictor for gains in verbal short-term memory (B=–1.57) and executive functions (B=–13.08).
- Lower age was a predictor for gains in verbal long-term memory (B=–0.06).
- Higher age was a predictor for gains in set-shifting (B=1.25).
- Higher education was a predictor for gains in verbal short-term memory (B=0.15) and attention (B=0.12).
ApoE4 Subgroup Analysis
Because the sample size of all participants tested for carrying the apoE4 allele was too small for us to include the apoE4 variable in the analysis of predictors, we used t tests to compare the change in scores (posttest–pretest) between carriers and noncarriers of the apoE4 allele for all outcomes. Even though the sample was small and not equally distributed (nine apoE4 carriers vs 26 noncarriers), t tests are robust against violations of their assumptions (Field, 2009).
At baseline, there were no significant differences between the carriers and noncarriers of apoE4 in any of the tested domains. The subgroup analysis revealed that noncarriers of apoE4 showed larger differences between pre- and posttraining in verbal long-term memory, t33=–2.38, P<0.05, and attention, t27=–2.47, P<0.05, than carriers (Figure 2).
The main finding of this study is that healthy older adults who participated in a structured cognitive training program showed a statistically significant improvement in the domain of verbal short-term memory compared with healthy older adults who participated in an unstructured brain jogging program and healthy older adults who received no cognitive training. Furthermore, on a descriptive level, a clear trend in favor of the structured cognitive training program was demonstrated in the domain of attention. We also discovered that improvement of cognitive functions in the tested domains after a structured cognitive training program can be predicted, to a large degree, by baseline performance on neuropsychological tests and, to a lesser degree, by the sociodemographic factors of age, sex, and education. In addition, subgroup analysis revealed that noncarriers of the apoE4 allele showed significantly more benefits in verbal long-term memory and attention after participating in cognitive training.
Our results are in line with many studies on cognitive training that have shown benefits in healthy older adults. Specifically, various studies have shown a positive effect of cognitive training on verbal memory (eg, Ball et al, 2002) and attention (eg, Mozolic et al, 2011). A recent meta-analysis of randomized controlled trials investigating the effects of cognitive training on healthy older adults demonstrated an overall effect for memory (effect size [Hedges’s g]=0.354) and attention (Hedges’s g=0.218) in favor of the intervention group (Chiu et al, 2017). These results are of fundamental importance in that attention is involved in most cognitive functions, such as memory and information processing, which are necessary for an independent and autonomous life (Posner, 2011). Furthermore, the decline of attention and verbal memory is part of the normal process of cognitive aging (Kramer and Madden, 2008).
Although we did not find a significant benefit of cognitive training on executive function/set-shifting, this is not unusual. The results of previous research on the success of cognitive training on executive functions are highly heterogeneous in terms of the subdomains investigated and the tests used (eg, Reijnders et al, 2013).
The fact that the structured training program was beneficial for our study participants is highly relevant to preventing cognitive aging because it demonstrates that the type of training makes a difference. Although our two programs differed in several aspects, we suspect that the structure of the training program was a crucial feature on which the benefits depended. This would be in line with the findings of Jaeggi et al (2008), who showed that the best effects were achieved using a structured working memory training program in a sample of healthy participants, and of Petrelli et al (2014), who showed that only a scientifically based, structured training program was beneficial for enhancing cognition compared with unspecific brain jogging and a wait list control group in a sample of nondemented patients with Parkinson disease.
The full rationale for this benefit is yet to be determined, but one main reason may be that structured cognitive training makes it easier for participants to integrate gained knowledge into established competence. Other reasons for the specific benefit of the structured program could be that the domains were specifically tailored to healthy older adults insofar as the training concentrated on the most vulnerable functions in this target group and that the contents from the psychoeducational part could be used for more efficient training. Although our study design does not permit clear conclusions, it does allow the statement that the type of training program used to increase cognition makes a difference and that structured training seems efficient to enhance cognition in healthy older adults.
Our results showed that low baseline scores on neuropsychological tests in the specific cognitive domains were predictive for improvements in these domains. It is important to note, however, that our study included a relatively high-functioning sample of older adults. It is possible that among these participants, the “lower high” can benefit more than the “higher high.” Also, if people realize their limitations in solving a task, they may be more motivated to participate in training, which could lead to more improvement in those participants with an initial low baseline performance. Previously Whitlock et al (2012), reported that the lower range of a high-functioning sample can benefit more from cognitive training, but Fairchild et al (2013) found that higher baseline scores were predictive for gains with cognitive training. This discrepancy merits further investigation.
According to Rebok et al (2013), the impact of sex on training-related cognitive outcomes is not well understood and is still underinvestigated. In the present study, being a woman was a predictor for gains in verbal short-term memory and executive functions/set-shifting, which is in line with the results of Rahe et al (2015b), who investigated participants with MCI. This supports the assumption of a concept of sex-specific plasticity that was introduced by Beinhoff et al (2008). Those authors concluded that the advantage in sex-specific cognitive domains of the healthy older adults might result in larger plasticity, particularly in verbal episodic memory in women versus visuospatial abilities in men.
Higher level of education was a significant predictor for improvement in verbal short-term memory and attention. This agrees with previous research suggesting that higher educational levels can be associated with higher benefits on memory measures (Langbaum et al, 2009). It is also possible that education not only represents the years of schooling that the participants had, but may also be a proxy variable for socioeconomic status, health practices, and even the willingness to learn new abilities over the whole life span (Krieger et al, 1997; Langbaum et al, 2009; Leigh and Fries, 1994), each affecting performance improvement due to cognitive training.
Subgroup analyses showed that the healthy older adults who were noncarriers of the apoE4 allele benefited significantly more in verbal long-term memory and attention tasks from cognitive training than did carriers of the apoE4 allele. The effects of apoE4 on cognition have been described not only for patients suffering from Alzheimer disease (Liu et al, 2013), but also for healthy older adults (Wisdom et al, 2011). However, to our knowledge, so far only one study has reported that apoE4 affected gains after a cognitive training program in patients with MCI (Binetti et al, 2013), and only one other study (Rahe et al, 2015a) has reported that apoE4 was predictive for gains in executive functions (an alternating letter verbal fluency task) induced by combined cognitive-physical training. Interpretation of our results must take into account the small sample size (nine apoE4 carriers vs 26 noncarriers). Further studies with larger sample sizes are needed to confirm our preliminary data, and therefore future research will need to unravel what role apoE4 plays in the biological mechanisms of cognitive training–induced cognitive plasticity.
Some limitations of our study need to be considered when interpreting the results. A first limitation is the fact that the conducted clinical trial was not registered. All future studies should be registered to ensure transparent working processes. Furthermore, only a part of the study (the comparison of the NEUROvitalis group vs the Mentally Fit group) was conducted in a randomized design; the control group was assessed separately. Although all three groups were similar with regard to age, sex, education, and cognitive baseline level, further studies with full randomization are needed to replicate our results.
A further limitation could be that the (total) study sample may have been biased. The baseline education of the participants was high in all of the groups, and our sample represents highly motivated and active healthy older people. However, this appears to be a more general problem that affects most intervention studies of healthy older people, as participating in a study is always voluntary, and it can be assumed that volunteers differ, at least in motivation (Oswald et al, 2006). Unfortunately, an overview of other activities in which the participants of all three groups might have attended during the study period was not obtained. Detailed protocols of further off-study activities, as performed, for example, by Graessel et al (2011) in patients with dementia, might help to estimate their possible influence and should be integrated in future studies.
In addition, as indicated here, our two training programs differed not only in structure, but also in the elements that were added to the cognitive task. NEUROvitalis, for example, included psychoeducation, whereas Mentally Fit included unspecific conversations. Because these topics were not registered, it is not possible to determine the degree to which the topics in the Mentally Fit conversations overlapped with the ones of psychoeducation. Accordingly, it is not possible to assign group differences to specific characteristics of the programs; other studies with training programs that differ in only single characteristics will have to be conducted to answer this question.
Another limitation of our study was its relatively small sample size: Power analysis indicated a power of only 42% to detect small interaction effects. However, the detection rate was 97% for medium interaction effects and 99% for strong interaction effects. Despite the small sample size, clear results were found in favor of the structured training, which emphasizes their validity.
The study also had limitations regarding the inclusion and exclusion criteria, as we screened hearing, vision, and psychiatric and neurologic disorders only with participants’ self-reports. Therefore, our sample might include participants who were not 100% healthy. Furthermore, we excluded participants with a suspected dementia according to a cognitive screening instrument (DemTect≤9 points) but did not exclude those with MCI. Still, it should be noted that only two participants in the control group showed a DemTect score of 12 points, which is the upper cutoff for MCI (DemTect 0–9 points=suspected dementia; 10–12=MCI; >12 points=cognitively healthy). No other participants scored in the range for MCI or dementia.
A methodological limitation is that no posttests were examined with those participants who were excluded or dropped out, and so an intention-to-treat-analysis could not be performed. However, because these participants did not differ from the analyzed participants in regard to age, sex, or education at baseline, a potential dropout bias seems unlikely.
This study does not include follow-up data. There is an ongoing debate on whether cognitive training effects remain after the intervention ends (Zelinski et al, 2011). Even though we focused on the specific characteristics of cognitive training and on possible predictors for its benefits rather than on long-term cognitive training effects, future studies should include follow-up examinations to address this important topic.
One question that cannot be answered in our present study is that of the transferability of the effects of cognitive training, as our training was a multicomponent intervention. Other study designs that use cognitive training programs targeting specific cognitive functions are necessary to shed light on this topic (for a discussion, see Simons et al, 2016). A recent meta-analysis by Kelly et al (2014), which included 25 randomized controlled trials, showed that transfer effects occurred more often when the training was adaptive than when it was not adaptive. By contrast, the ACTIVE trial showed only limited evidence for the transfer of training effects from one skill to another; in that large intervention study, improvements were mostly limited to the trained activities (Ball et al, 2013; Willis and Caskie, 2013). In summary, the present data reveal inconsistent results on transfer and long-term training effects (Ballesteros et al, 2015), and therefore it is important to address the question of transferability of cognitive training effects in future studies.
The present study contributes to the previous research on effectiveness in cognitive training by showing that the structured NEUROvitalis training program has the potential to enhance cognitive functions in healthy older adults. One strength of this study is the use of an active control group (the Mentally Fit intervention group) next to a passive (no intervention) control group to minimize the occurrence of unspecific effects, such as increased social activity. Finally, the randomized design (of the structured cognitive training program and the unstructured brain jogging program) and the fact that adherence to the intervention was monitored are other strengths of the study.
This study showed that a structured cognitive training program could be used to improve cognitive functions compared not only with a nonintervention control group, but also with an unstructured brain jogging program. Thus, the study indicates that healthy older adults who are afraid of cognitive decline due to their advanced age should most likely participate in structured cognitive training programs in order to train and strengthen their cognitive functions. Moreover, low baseline performance on neuropsychological tests could be identified as a significant predictor for training gains in all of the investigated domains, and noncarriers of the apoE4 allele could be identified to benefit significantly more from cognitive training in the domains of long-term verbal memory and attention. The identification of predictors of cognitive training is highly important to identify responders’ characteristics, which would help to provide cognitive training programs that are matched to specific individuals in terms of, for example, training difficulty to improve the overall effectiveness of cognitive training (Fairchild et al, 2013). Future studies will have to investigate further the meaning of predictors for the success of cognitive training in healthy older adults.
The authors thank all of the participants for their interest in this study. The authors also thank Stephanie Kaesberg, PhD, and Annette Petrelli, PhD, at the University of Vechta, and Julia Meyer, PhD, at the University of Osnabrück, for assessing part of the data.
Bahar-Fuchs A, Clare L, Woods B. 2013. Cognitive training
and cognitive rehabilitation for mild to moderate Alzheimer's disease and vascular dementia. Cochrane Database Syst Rev. doi:10.1002/14651858.CD003260.pub2
Ball K, Berch DB, Helmers KF, et al. 2002. Effects of cognitive training
interventions with older adults: a randomized controlled trial. JAMA. 288:2271–2281.
Ball K, Ross LA, Roth DL, et al. 2013. Speed of processing training in the ACTIVE study: how much is needed and who benefits? J Aging Health. 25 (suppl):65S–84S. doi:10.1177/0898264312470167
Baller G. 2003. Kognitives Training: Ein sechswöchiges Übungsprogramm für Senioren zur Verbesserung der Hirnleistung [Cognitive Training
: A Six Week Training Program for Older Adults to Improve Brain Activity]. Bad Honnef, Germany: Hippocampus Verlag KG.
Baller G, Kalbe E, Kaesbert S, et al. 2009. NEUROvitalis Ein Neuropsychologisches Gruppenprogramm zur Förderung der Geistigen Leistungsfähigkeit [NEUROvitalis A Neuropsychological Group Program to Improve Cognitive Abilities]. Cologne, Germany: ProLog.
Ballesteros S, Kraft E, Santana S, et al. 2015. Maintaining older brain functionality: a targeted review. Neurosci Biobehav Rev. 55:453–477.
Bamidis PD, Fissler P, Papageorgiou SG, et al. 2015. Gains in cognition through combined cognitive and physical training: the role of training dosage and severity of neurocognitive disorder. Front Aging Neurosci. 7:152. doi:10.3389/fnagi.2015.00152
Bamidis PD, Vivas AB, Styliadis C, et al. 2014. A review of physical and cognitive interventions in aging. Neurosci Biobehav Rev. 44:206–220.
Barbian G, Lutz M, Thomas M. 1995. Lebensreise Ein Generationenspiel zum Entdecken der eigenen Lebensgeschichte im gemeinsamen Spiel und Gespräch [Life Journey A Game to Explore the Story of Your Life by Playing Games and Talks]. Trier, Germany: Rudolf Günther Verlag.
Beck AT, Ward CH, Mendelson M, et al. 1961. An inventory for measuring depression. Arch Gen Psychiatry. 4:561–571.
Beinhoff U, Tumani H, Brettschneider J, et al. 2008. Gender-specificities in Alzheimer's disease and mild cognitive impairment. J Neurol. 255:117–122. doi:10.1007/s00415-008-0726-9
Binetti G, Moretti DV, Scalvini C, et al. 2013. Predictors of comprehensive stimulation program efficacy in patients with cognitive impairment. Clinical practice recommendations Int J Geriatr Psychiatry. 28:26–33. doi:10.1002/gps.3785
Blazer DG, Yaffe K, Liverman CT. 2015. Cognitive Aging: Progress in Understanding and Opportunities for Action. Washington, DC: National Academies Press.
Boos A, Friese A, Lindenberg-Kaiser M, et al. 2007. Ganzheitliches Gedächtnistraining Spannende Themenstunden für die geistige Aktivierung von Gruppen mit geübten Teilnehmern
[Overall Memory Training Exciting Themes for Cognitive Activation of Groups With Experienced Participants]. Allendorf, Germany: Verlag Susanne Gassen.
Cheng Y, Wu W, Feng W, et al. 2012. The effects of multi-domain versus single-domain cognitive training
in non-demented older people: a randomized controlled trial. BMC Med. 10:30. doi:10.1186/1741-7015-10-30
Chiu HL, Chu H, Tsai J, et al. 2017. The effect of cognitive-based training for the healthy older people: a meta-analysis of randomized controlled trials. PloS One. 12:e0176742. doi:10.1371/journal.pone.0176742
Deary IJ, Corley J, Gow AJ, et al. 2009. Age-associated cognitive decline. Br Med Bull. 92:135–152. doi:10.1093/bmb/ldp033
Dorfman CR, Ager CL. 1989. Memory and memory training: some treatment implications for use with the well elderly. Phys Occup Ther Geriatr. 7:21–42. doi.org/10.1080/J148v07n03_03
Erdfelder E, Faul F, Buchner A. 1996. GPOWER: a general power analysis program. Behav Res Methods. 28:1–11. doi:10.3758/BF03203630
Fairchild J, Friedman L, Rosen A, et al. 2013. Which older adults maintain benefit from cognitive training
? Use of signal detection methods to identify long-term treatment gains. Int Psychogeriatr. 25:607–616. doi:10.1017/S1041610212002049
Field A. 2009. Discovering Statistics Using SPSS, 3rd ed. London, United Kingdom: Sage Publications.
Graessel E, Stemmer R, Eichenseer B, et al. 2011. Non-pharmacological, multicomponent group therapy in patients with degenerative dementia: a 12-month randomized, controlled trial. BMC Med. 9:129. doi.org/10.1186/1741-7015-9-129
Halbach A. 1998. Gedächtnistraining in 10 Themen Band 2 [Memory training in 10 themes Band 2]. Stuttgart, Germany: Memo Verlag.
Jaeggi SM, Buschkuehl M, Jonides J, et al. 2008. Improving fluid intelligence with training on working memory. PNAS. 105:6829–6833. doi:10.1073/pnas.0801268105
Kelly ME, Loughrey D, Lawlor BA, et al. 2014. The impact of exercise on the cognitive functioning of healthy older adults: a systematic review and meta-analysis. Ageing Res Rev. 16:12–31.
Kessler J, Calabrese P, Kalbe E, et al. 2000. DemTect: a new screening method to support diagnosis of dementia. Psycho. 26:343–347.
Kessler J, Calabrese P, Kalbe E. 2010. DemTect-B: a parallel test version to the cognitive screening instrument Dem Tect-A. Fortschr Neurol Psychiatr. 78:532–535. doi:10.1055/s-0029-1245452
Kramer A, Madden DCraik FIM, Salthouse TA. 2008. Attention. The Handbook of Aging and Cognition. New York, New York: Psychology Press; 189–249.
Krieger N, Williams DR, Moss NE. 1997. Measuring social class in US public health research: concepts, methodologies, and guidelines. Annu Rev Public Health. 18:341–378. doi:10.1146/annurev.publhealth.18.1.341
Lampit A, Hallock H, Valenzuela M. 2014. Computerized cognitive training
in cognitively healthy older adults: a systematic review and meta-analysis of effect modifiers. PLoS Med. 11:e1001756. doi:10.1371/journal.pmed.1001756
Langbaum JB, Rebok GW, Bandeen-Roche K, et al. 2009. Predicting memory training response patterns: results from ACTIVE. J Gerontol B Psychol Sci Soc Sci. 64:14–23. doi:10.1093/geronb/gbn026
Leigh JP, Fries JF. 1994. Education, gender, and the compression of morbidity. Int J Aging Hum Dev. 39:233–246. doi:10.2190/XQXR-UTGP-WA8X-9FQJ
Liu C-C, Kanekiyo T, Xu H, et al. 2013. Apolipoprotein E and Alzheimer disease: risk, mechanisms and therapy. Nat Rev Neurol. 9:106–118. doi:10.1038/nrneurol.2012.263
Mozolic JL, Long AB, Morgan AR, et al. 2011. A cognitive training
intervention improves modality-specific attention in a randomized controlled trial of healthy older adults. Neurobiol Aging. 32:655–668. doi:10.1016/j.neurobiolaging.2009.04.013
Normann U. 1996. Heiteres Gedächtnistraining: Wortspielereien im Großdruck nach der Stengel-Methode [Funny Memory Training: Game of Words in Big Printed Letters Following the Stengel Method]. Stuttgart, Germany: Memo Verlag.
Oppolzer U. 1998. Hirntraining mit ganzheitlichem Ansatz: Grundlagen, Anregungen und Trainingsmaterial für Gruppenleiter und Dozenten [Brain Training With Whole-Brain Approach: Basis, Ideas and Training Material for Teachers]. Dortmund, Germany: Verlag Modernes Lernen.
Oswald WD, Gunzelmann T, Rupprecht R, et al. 2006. Differential effects of single versus combined cognitive and physical training with older adults: the SimA study in a 5-year perspective. Eur J Ageing. 3:179–192. doi:10.1007/s10433-006-0035-z
Petrelli A, Kaesberg S, Barbe MT, et al. 2014. Effects of cognitive training
in Parkinson's disease: a randomized controlled trial. Parkinsonism Relat Disord. 20:1196–1202. doi:10.1016/j.parkreldis.2014.08.023
Petrelli A, Kaesberg S, Barbe MT, et al. 2015. Cognitive training
in Parkinson’s disease reduces cognitive decline in the long term. Eur J Neurol. 22:640–647.
Posner MI. 2011. Cognitive Neuroscience of Attention, 2nd ed. New York, New York: Guilford Press.
Rahe J, Becker J, Fink GR, et al. 2015a. Cognitive training
with and without additional physical activity in healthy older adults: cognitive effects, neurobiological mechanisms, and prediction of training success. Front Aging Neurosci. 7:187. doi:10.3389/fnagi.2015.00187
Rahe J, Liesk J, Rosen JB, et al. 2015b. Sex differences in cognitive training
effects of patients with amnestic mild cognitive impairment. Neuropsychol Dev Cogn B Aging Neuropsychol Cogn. 22:620–638. doi:10.1080/13825585.2015.1028883
Rebok GW, Langbaum JB, Jones RN, et al. 2013. Memory training in the ACTIVE study: how much is needed and who benefits? J Aging Health. 25 (suppl):21S–42S. doi:10.1177/0898264312461937
Reijnders J, van Heugten C, van Boxtel M. 2013. Cognitive interventions in healthy older adults and people with mild cognitive impairment: a systematic review. Ageing Res Rev. 12:263–275. doi:10.1016/j.arr.2012.07.003
Reitan RM. 1958. Validity of the Trail Making Test as an indicator of organic brain damage. Percept Mot Skills. 8:271–276. doi:10.2466/PMS.8.7.271-276
Reuter-Lorenz PA, Park DC. 2014. How does it STAC up? Revisiting the scaffolding theory of aging and cognition. Neuropsychol Rev. 24:355–370. doi:10.1007/s11065-014-9270-9
Schretlen D, Bobholz JH, Brandt J. 1996. Development and psychometric properties of the Brief Test of Attention. Clin Neuropsychol. 10:80–89. doi:10.1080/13854049608406666
Schulz KF, Altman DG, Moher D, et al. 2010. CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials. BMJ. 340:c332. doi:10.1136/bmj.c332
Simons DJ, Boot WR, Charness N, et al. 2016. Do “brain-training” programs work? Psychol Sci Public Interest. 17:103–186.
Stengel F. 1997. Heitere Gedächtnisspiele 3 Training zur geistigen Konzentration [Funny Memory Games 3 Training to Improve Concentration]. Stuttgart, Germany: Memo Verlag.
Tanklage E. 2009. Gedächtnistraining für Seniorengruppen: 24 unterhaltsame Stundenfolgen für Gruppenleitungen
[Memory Training for Seniors: 24 Funny Hours for Group Sessions]. Weinheim, Germany: Juventa.
Whitlock LA, McLaughlin AC, Allaire JC. 2012. Individual differences in response to cognitive training
: using a multi-modal, attentionally demanding game-based intervention for older adults. Comput Human Behav. 28:1091–1096. doi:10.1016/j.chb.2012.01.012
Willis SL, Caskie GI. 2013. Reasoning training in the ACTIVE study: how much is needed and who benefits? J Aging Health. 25 (suppl):43S–64S. doi:10.1177/0898264313503987
Willis SL, Tennstedt SL, Marsiske M, et al. 2006. Long-term effects of cognitive training
on everyday functional outcomes in older adults. JAMA. 296:2805–2814.
Wisdom NM, Callahan JL, Hawkins KA. 2011. The effects of apolipoprotein E on non-impaired cognitive functioning: a meta-analysis. Neurobiol Aging. 32:63–74. doi:10.1016/j.neurobiolaging.2009.02.003
Zelinski EM, Spina LM, Yaffe K, et al. 2011. Improvement in memory with plasticity-based adaptive cognitive training
: results of the 3-month follow-up. J Am Geriatr Soc. 59:258–265.
Zhu X, Yin S, Lang M, et al. 2016. The more the better? A meta-analysis on effects of combined cognitive and physical intervention on cognition in healthy older adults. Ageing Res Rev. 31:67–79.
Zinke K, Zeintl M, Rose NS, et al. 2014. Working memory training and transfer in older adults: effects of age, baseline performance, and training gains. Dev Psychol. 50:304–315.