Postural control and cognitive task performance in healthy participants while balancing on different support-surface configurations

Abstract

Postural control during normal upright stance in humans is a well-learned task. Hence, it has often been argued that it requires very little attention. However, many studies have recently shown that postural control is modified when a cognitive task is executed simultaneously especially in the elderly and in the presence of pathology. This study examined postural control modifications when a cognitive task of varying difficulty levels is added. Postural stance difficulty was also varied. Results from this study suggest that a generalized capacity interference may occur due to the larger interference found with the addition of a cognitive task in the more novel and difficult postural task. Because the performance of the cognitive task was tapered by a speed-difficulty trade-off, it was not possible to determine whether a change in the level of difficulty of the cognitive task occurred and if it would produce larger dual-task interference.

Introduction

Postural control during normal upright stance in humans is a well-learned task. Hence, it has often been argued that it requires very little attention [1], [2]. Postural control is subserved by numerous neural pathways at spinal and supraspinal levels that constitute elementary reflexes and initially learned synergies which form the basis for fast responses to body perturbations [3]. These reflexes and synergies provide a continuous parametric control of gain and phase of feedback sensorimotor loops directed at maintaining a certain state of equilibrium [4]. Therefore, this lower level mode of control is usually regarded as independent of attention demands because it requires only a minimum of computational activity [5]. However, recent research has shown that dual-task paradigm involving increased cognitive demand can modify postural control [6], [7], [8], [9], [10].

The involvement of cognitive processes in the control of posture first became apparent when considering the role of feedforward control in adaptation to motor goals. It was shown that prior knowledge affected both the timing of anticipatory postural adjustments when comparing unexpected to self-initiated arm movements [11], [12], [13], as well as the magnitude of the postural responses to externally induced body perturbations by modifying ‘central set’ based on prior experience [14]. Still, it was hypothesised that cognitive influence on postural control was discontinuous, that is, during short periods of adaptation to new equilibrium states, e.g. during an alteration of support-surface configuration [15], [16]. Even in such instances, because of the low attentional demands needed to maintain postural control using pre-structured synergies, marked vulnerability of postural activity to cognitive task performance on the basis of central capacity interference was not likely to occur [17], [18].

However, in recent years, the automaticity of postural control has been challenged. Kerr et al. [19] found that the performance of a cognitive (visuo-spatial) task was modified when participants were asked to simultaneously execute a difficult balance task. Also, changes in postural sway have been found when participants are asked to execute a cognitive task indicating that attention may play a role in the control of posture [8], [20], [21], [22]. Dependency on attentional processes seems even more apparent when the central nervous system (CNS) is impaired such as in elderly participants [6], [7], [8], [23], [24], [25], [26] and in the presence of pathology [9], [10], [22].

If we consider that motor control and cognitive processing are carried out in parallel by using time-sharing strategies, the difficulty and novelty of the tasks will have a great impact on how well both types of information processing can be performed simultaneously [27]. If the level of difficulty of one of the two tasks is increased, it may be reflected by a reduction in the performance quality of the other task. Since shoulder width stance is a well-learned skill it should require little attention [27]. On the other hand, if the postural task is more difficult, we could hypothesise that the interference caused by simultaneous cognitive processing, may be greater since the primary task will require more attention. Inversely, if the cognitive task difficulty is increased, less attention resources may be available for postural control and an increased interference may also occur. To test the validity of these hypotheses, this study addresses the question if and to what extent postural control can be influenced by cognitive task performance even in healthy young adults balancing on different support-surface configurations.

In addition to quiet upright standing on a firm and flat support surface (shoulder width stance), we examined standing on two other support-surface configurations in order to interfere with the efficacy of commonly employed postural strategies. Firstly, participants were requested to balance on a pair of seesaws, thus complicating the utilisation of vertical ground reaction forces through ankle torque generation to control antero-posterior body sway [5], [18]. This manipulation was believed to require only a change in parameterisation (timing and gain) of well-developed synergies mainly in the sagittal plane. In order to increase the level of difficulty and to add a novelty aspect, we asked participants to also stand in a tandem stance on the same seesaws, thus completely eliminating the intrinsic mechanical stability of lateral balance which is normally provided by double-limb support in the frontal plane. By having them stand on the seesaws, we hoped to also complicate the control of antero-posterior sway by reducing the efficacy of ankle mechanisms in the sagittal plane. However, in a tandem position, bipedal stability was now available in this plane. Hence, tandem seesaw stance was primarily expected to induce a change in balance strategy towards the generation of high frequency ankle torques working in the frontal plane. Because this control mechanism is less practised in daily activities, a clear dual-task effect on lateral sway control was predicted for this task.

Three levels of difficulty of the cognitive task were chosen. The Stroop task was selected because its performance requires a considerable amount of attention even after many repetitions and because it comprises three discrete levels of complexity [28]. It was assumed that the three Stroop tasks, with increasing level of difficulty, would demand an increasing amount of attentional capacity and, therefore, would increasingly interfere with postural control especially during tandem seesaw stance.