Author(s): PRABHU, N.M., MATKE, M., LEHMANN, N., AYE, N., ZIEGLER, G., TAUBERT, M., Institution: OTTO-VON-GUERICKE UNIVERSITÄT MAGDEBURG, Country: GERMANY, Abstract-ID: 2446

INTRODUCTION:
Positive effects of exercise on cognition are thought to be especially strong at either ends of the life span spectrum [1]. Particularly in case of older adults approaching physical and cognitive decline, optimisation of exercise induced benefits is necessary [2]. One proposed approach is to match the difficulty level of the task being trained to the individual’s abilities and applying this optimised ratio during training [3]. Therefore, we hypothesized that optimal training conditions would induce higher learning gains and enhance transfer effects onto untrained motor and cognitive tasks.
METHODS:
We conducted a randomized, single-blinded, 6-week dynamic balance training (DBT) intervention [4] optimised to the participants individual balance ability [3] (n=30, age range 60-80 yrs). Balance ability was assessed using 6-different levels of task difficulties ranging from level 0 (highest) to level 5 (lowest). Training was formulated to emulate either overload, underload or optimal load conditions. Cognitive (memory and executive) and near-far motor transfer (untrained variations of DBT and Wii) were investigated half-way through training duration (mid) and post training (post). Statistical analyses were conducted using repeated measures ANOVA with simultaneous component analysis (RM-ASCA+), a method combining linear mixed models with PCA [5]. Using the optimal group as the reference condition, group differences were inferred from non-overlapping 95% bootstrap intervals of the factor scores [5].
RESULTS:
RM-ASCA+ revealed higher learning gains after the first training session in the optimal training group (95% CI [0.17, 1.44]) compared to both overload (95%CI [-2.46, -0.07]) and underload conditions (95%CI [-2.08, 0.03]). Similarly, higher learning gains after second (optimal vs underload: 95%CI [-1.76, 0.089]) and fifth training sessions (optimal vs overload: 95%CI [-1.83, 0.22]) were detected. This advantage of optimal training load was more pronounced at higher task difficulty levels as group differences were driven by higher factor loadings at level 1, 2 and 5. Furthermore, optimal load (95%CI [0.34, 1.01]) induced higher transfer effects mid- (underload: 95%CI [-1.29, 0.47], overload: 95%CI [-1.64, -0.3]) and post-intervention (underload: 95%CI [-1.46, -0.04], overload: 95%CI [-2.01, -0.34]). These differences were driven by higher factor loadings for motor transfer and executive tasks.
CONCLUSION:
Our results show that benefits of optimising practice conditions in old age are not just limited to greater learning gains on the trained task, but are also reflected as performance improvements in transfer tasks. These findings further highlight the underlying neural mechanisms involved that enable transfer across domains.
[1] Tomporowski & Pesce, Psychol Bull., 2019
[2] Ludyga et al., Nat Hum Behav., 2020
[3] Guadagnoli & Lee, J Mot Behav., 2004
[4] Taubert et al., J Neurosci., 2010
[5] Jarmund et al., Front Mol Biosci., 2022