Cognitive-Postural Multitasking Training in Older Adults - Effects of Input-Output Modality Mappings on Cognitive Performance and Postural Control.

Older adults exhibit impaired cognitive and balance performance, particularly under multi-task conditions, which can be improved through training. Compatibility of modality mappings in cognitive tasks (i.e., match between stimulus modality and anticipated sensory effects of motor responses), modulates physical and cognitive dual-task costs. However, the effects of modality specific training programs have not been evaluated yet. Here, we tested the effects of cognitive-postural multi-tasking training on the ability to coordinate task mappings under high postural demands in healthy older adults. Twenty-one adults aged 65-85 years were assigned to one of two groups. While group 1 performed cognitive-postural triple-task training with compatible modality mappings (i.e., visual-manual and auditory-vocal dual n-back tasks), group 2 performed the same tasks with incompatible modality mappings (i.e., visual-vocal and auditory-manual n-back tasks). Throughout the 6-weeks balance training intervention, working-memory load was gradually increased while base-of-support was reduced. Before training (T0), after a 6-week passive control period (T1), and immediately after the intervention (T2), participants performed spatial dual one-back tasks in semi-tandem stance position. Our results indicate improved working-memory performance and reduced dual-task costs for both groups after the passive control period, but no training-specific performance gains. Furthermore, balance performance did not improve in response to training. Notably, the cohort demonstrated meaningful interindividual variability in training responses. Our findings raise questions about practice effects and age-related heterogeneity of training responses following cognitive-motor training. Following multi-modal balance training, neither compatible nor incompatible modality mappings had an impact on the observed outcomes.