This randomized controlled trial was designed to test whether a twelve-week strength and balance exercise regimen, that includes a dance video game as an additional cognitive element, would lead to greater changes in measures of gait performance and fear of falling, compared to strength and balance exercise alone. Although both groups attained improvements in gait performance and were able to reduce their concerns about falling, the results suggest positive interaction effects in favor of the dance video game group. The finding of this study supports the notion that it is advantageous to combine physical and cognitive training into clinical practice. The combination seems to have a positive influence on older adults walking abilities under dual task conditions in comparison to more traditional exercise forms
The most prominent differences between the training groups were observable in the gait analysis. The CG demonstrated significant positive within-group changes of several spatio-temporal parameters, however, merely in the single task condition and at preferred gait speed (normal). Furthermore, this group exhibited a gain in velocity in the normalcog condition. This merely confirms findings from a systematic review that a strength and balance exercise regimen is able to preserve or improve walking abilities, even in advanced age
. The goal of this study, however, was to improve walking behavior under dual task conditions. The results of previous studies with similar groups, which were performing progressive machine-driven resistance training complemented with functional balance exercises, revealed no improvement of performance under attention demanding circumstances; e.g. no changes in the dual task costs of walking
[50, 51]. Daily activities pose high cognitive demands and safe walking should be practicable also under cognitive distractive or otherwise challenging conditions. The results of the DG show significant positive within-group differences for most gait parameters also in the dual task conditions normalcog and fastcog, thus confirming findings from previous pilot studies with similar results for dual task related costs
[27, 30]. Furthermore, significant between-group differences in the dual task condition fastcog were observed for gait velocity and single support time in favor of the DG. Fastcog is the condition with the most challenging motor and cognitive demands. In the present study the positive effect on DTC of gait, represented by the decrease in Δ DTC values in the DG (Table
4), may be attributed to the additional input provided by the dance video game. Thus, this substantiates the hypothesis that an additional cognitive challenge should be preferably part of a training program aiming to improve physical functioning in older adults, especially under dual task conditions. Unfortunately, how gait under these conditions should be improved has not yet been well-studied in general
 and this study is one of the first that shows that an improvement in dual task walking with an exercise intervention supplemented by a video game is achievable.
In the FPA test both groups revealed a more accurate foot placement in M-L direction over all the walking conditions, however, only the DG manifested significant within-group differences after the intervention. The better performance may be in part attributable to improvements in walking and balance skills gained by the strength and balance exercises. A higher postural balance confidence during swing phase of the gait cycle has possibly enabled a more accurate targeting. However, a more efficient movement planning and a possible change in visual scanning of the walking path have possibly led to the better performance in the FPA test in favor of the DG. Interactive video games, like the dance video game used in this study, require precise visuo-motor control, that is to focus attention on the screen and the concurrent execution of controlled body movement and the regulation of postural control. Interestingly in this context is that expert action video game players were found to have an improved spatial distribution and resolution of visual attention, a more efficient visual attention over time and were able to attend a higher number of objects simultaneously compared to non-players
[53, 54], thus allowing a better allocation of the attentional resources over a visuo-motor task.
Interestingly, in both exercise groups the mean distance error in A-P direction increased after the intervention. Participants were able to navigate quicker through the test path thereby controlling their M-L direction walking deviation, however, suffered the loss of accuracy in mean distance error in A-P direction. The higher inaccuracy in the A-P direction may be in part explained by the higher waking velocity in the second test. It can be suggested, that participants gave more priority to their walking performance (greater velocity, larger steps) and their navigation towards the target in M-L direction, so that their foot placement accuracy in A-P direction decreased (speed-accuracy tradeoff)
. The quality evaluation, however, shows in general a qualitative better performance in the second FPA test with less shoe contacts with the targets (Table
The reason to use a dance video game or video games in general, is mainly based on the findings of a systematic review
. It is, however, also related to the numerous advantages attributed to such a tool
. As known from the principles of motor learning, repetition is important for both motor learning and the cortical changes that initiate it
. The repeated practice must be linked to incremental success at some task or goal. A computerized intervention like the dance video game constitutes a powerful tool to provide participant repetitive practice, feedback about performance and motivation to endure practice
. In addition, it can be adapted based on the individual participant’s baseline motor performance and be progressively augmented in task difficulty. Further, the addition of a challenging video game has the potential to engage people who otherwise would lack of interest to participate in a physical exercise regimen. Especially in the older population it is difficult to maintain high adherence to training programs
. The participants of the present study showed excellent compliance rates. The losses related to low exercise compliance or low motivation (n = 4) in the DG were caused by animosities between the participants of one training group and in part by not perceiving any changes in performance level at the half of the study. The reasons for discontinuation of training were not because of rejection of the dance video game per se. The DG members were motivated by the additional playing of the video game at the end of every training session. The CG members were motivated by the assurance that after the end of the intervention they had the opportunity to include the dance video game in their exercise program as well. In both hostels, the training sessions with the additional dance video game were pursued also after the study ended.
The high acceptance of the dance video game used in our study seems at variance with reports of elderly being rather skeptical towards using commercially available games in a hospital setting
. We think that this is partly explainable due to the modifications made to the original dance video game free-ware StepMania. The information on the screen was reduced to a minimum and the music was chosen according participants’ taste. In general, commercially available video games are often not adapted to the needs and preferences of older adults, since they are designed for children and young adults. The games are not easy to comprehend and the screens are flashing. This might be one of the reasons why some commercially available video games are rather disliked by older adults
Limitations of the study
The present study contains some limitations that have to be discussed. A limitation of the FPA test is that different shoe sizes of the participants are not accounted for. The further development of the FPA test protocol should consider foot size by using different sized foam targets. We assume that a subject with small feet has more free space in the target area to place his/her foot before touching the border of the target resulting in a potentially higher risk of becoming variable in the accuracy performance. On the other hand the larger the foot the smaller the free space between foot and border of the target. A subject with large feet will not have a comparable amount of potential variability of distance errors as the person with small feet.
An obvious limitation of our study is the rather small sample size. This study, therefore, only reveals first estimates for these measures and warrants further research in larger populations. When evaluating the validity of a study it is important to consider both the clinical and statistical significance of the findings
. Studies that claim clinical relevance may lack sufficient statistical significance to make meaningful statements or, conversely, may lack practicality despite showing a statistically significant difference in treatment options. Researchers and clinicians should not focus on small P-values alone to decide whether a treatment is clinically useful; it is necessary to also consider the magnitude(s) of treatment differences and the power of the study
. Encouraging in this context is the observation that the majority of the between groups comparisons show medium or medium-to-high magnitude(s) of treatment differences. This in mind, the relationship between physical and cognitive training research and its effect on gait in elderly individuals requires further exploration. Future adequately powered studies with similar populations should, therefore, be performed to substantiate our assumption and findings.
The suggested link between the observed improvement in the physical tests after the intervention and influences on cognitive processes in the brain is as of yet still speculative. A necessary next step would be to investigate the isolated effects of the video game on measures of cognitive functioning. Since improvements were observable in physical performance under attention demanding circumstances it seems plausible to hypothesize that these changes may rely, at least in part, on functional or even structural changes in the brain. A recently published study protocol
 might be able to provide some insights on this topic.