Sway regularity and sway activity in older adults' upright stance are differentially affected by dual task

Age-related changes in postural control are attributed to visual, vestibular and proprioceptive dysfunctions, muscle weakness, and reduced availability of neural resources required for efficient balance control. Concurrent performance of complex cognitive tasks while standing or walking is expected to increase balance instability due to under-recruitment of brain resources and insufficient allocation of attention to the postural task. Both balance instability and attentional control of movements can, nonetheless, be determined from the center of pressure (CoP) measurements by examining the effects of dual-task on the amount of sway activity (as measured by CoP velocity - Vcop) and the statistical regularity of the CoP trajectory (the wavelet entropy of the signal - WEcop). The abovementioned characteristics were examined in the present study in a group of 13 healthy older adults. The task involved maintaining Romberg stance for 25 s with or without performing an attention demanding cognitive tasks (word memorization or mathematical counting). A linear mixed-model study was designed to analyze the extent to which sway activity can predict sway regularity. Findings from the present study showed that, on average, Vcop and WEcop were positively correlated (p = 0.014), suggesting that older individuals who exhibited greater amounts of sway (i.e., higher Vcop) also increased sway irregularity of the posturogram - as evidenced by a higher level of wavelet entropy of the CoP trajectory. Nevertheless, results of the linear mixed model showed that significant positive associations between Vcop and WEcop were found only in dual task (R ≥ 0.67, p ≤ 0.012). Furthermore, dual-task effects (% change in performance) on both sway characteristics were not significant (p > 0.1), suggesting that none of the attention demanding cognitive tasks used in the present study was sufficient to divert a critical amount of attentional resources from the postural task. Finally, performance of the mathematical counting (but not the word memorization) task was deteriorated from sitting to standing, however this effect was marginal (p = 0.075). Taken together, we proposed that while dual task could hinder balance control, postural stability may still be maintained by allocating more attentional resources to the postural task and reducing automatized control.