Neural Correlates of Balance Skill Learning in Young and Older Individuals: A Systematic Review and Meta-analysis

Abstract

Background Despite the increasing number of research studies examining the effects of age on the control of posture, the number of annual fall-related injuries and deaths continues to increase. A better understanding of how old age affects the neural mechanisms of postural control and how countermeasures such as balance training could improve the neural control of posture to reduce falls in older individuals is therefore necessary. The aim of this review is to determine the effects of age on the neural correlates of balance skill learning measured during static (standing) and dynamic (walking) balance tasks in healthy individuals. Methods We determined the effects of acute (1–3 sessions) and chronic (> 3 sessions) balance skill training on balance in the trained and in untrained, transfer balance tasks through a systematic review and quantified these effects by robust variance estimation meta-analysis in combination with meta-regression. We systematically searched PubMed, Web of Science, and Cochrane databases. Balance performance and neural plasticity outcomes were extracted and included in the systematic synthesis and meta-analysis. Results Forty-two studies (n = 622 young, n = 699 older individuals) were included in the systematic synthesis. Seventeen studies with 508 in-analysis participants were eligible for a meta-analysis. The overall analysis revealed that acute and chronic balance training had a large effect on the neural correlates of balance skill learning in the two age groups combined (g = 0.79, p < 0.01). Both age groups similarly improved balance skill performance in 1–3 training sessions and showed little further improvements with additional sessions. Improvements in balance performance mainly occurred in the trained and less so in the non-trained (i.e., transfer) balance tasks. The systematic synthesis and meta-analysis suggested little correspondence between improved balance skills and changes in spinal, cortical, and corticospinal excitability measures in the two age groups and between the time courses of changes in balance skills and neural correlates. Conclusions Balance skill learning and the accompanying neural adaptations occur rapidly and independently of age with little to no training dose-dependence or correspondence between behavioral and neural adaptations. Of the five types of neural correlates examined, changes in only spinal excitability seemed to differ between age groups. However, age or training dose in terms of duration did not moderate the effects of balance training on the changes in any of the neural correlates. The behavioral and neural mechanisms of strong task-specificity and the time course of skill retention remain unclear and require further studies in young and older individuals. Registration: PROSPERO registration number: CRD42022349573.

Cortical resting-state activities may predict anodal tDCS-induced facilitation of dynamic balance training

“…A transfer of dynamic balance training on static balance tasks could not be revealed. This is in accordance with a recent systematic review and meta-analysis by (Bakker et al, 2024) who also reported specificity of balance training with only minimal or no transfer to non-trained motor tasks.…”

“…Besides the sensorimotor cortex (particularly the leg area of the primary motor cortex, M1) we alternatively applied a-tDCS to the left DLPFC to examine which target region was more effective in facilitating DBT training relative to a sham condition potential transfer effects to static balance tasks. Results of a recent meta-analysis (Bakker et al, 2024) reported improvement in untrained tasks is minimal and not substantial. We assumed that the application of a-tDCS can facilitate changes in brain networks in targeted brain areas to not only improve motor execution but also top-down executive control.…”

“…A recent study by (Bakker et al, 2024) demonstrated that improvement in transfer balance tasks is more noticeable when the training task aligns with the transfer task. As a result, choosing similar training and transfer test tasks could reveal better learning transfer.…”

In older adults resting-state alpha power is associated with stronger effects of anodal tDCS over prefrontal cortex on dynamic balance

“…Therefore, we highlight the strong association between cognition and more complex balance tasks such as dynamic balance tasks (Divandari et al, 2023). Additionally, as reported by a recent systematic review and metaanalysis, balance training leads to improvement in transfer tasks almost only when the trained task aligns with the transfer task (Bakker et al, 2024).…”