Increasing cognitive load with increasing balance challenge: recipe for catastrophe

The variety of sometimes contradictory results of studies of the impact of secondary cognitive tasks on postural balance may be attributed to the heterogeneity of balance challenges and tasks deployed and frequent lack of quantitative comparability of tasks. We deployed a wide range of quantitatively graded difficulties of both balance challenge and cognitive tasking to obtain an overview of the spectrum of their interactions in a multi-tasking situation. A differential comparison of the effects of verbally versus spatially loaded tasks, balanced for difficulty, was made and unlike any other study, we contrived to incorporate falls as an experimental variable. In the first study subjects stood in tandem on beams of either 2, 3 and 6 cm or 3, 6 and 8 cm width (according to 'best performance' ability) while performing mental verbal or spatial 'Stroop' tasks. The design was a between groups (sixteen subjects each) comparison (to reduce learning effect) of sway, fall rate and task error, balanced for order. Measurements were taken of centre of pressure, sway velocity at the hip and head displacement. For any beam width there were no within-subject correlations between sway magnitudes and frequency of falls. Spatial task errors increased with balance challenge (hence with magnitude of sway) but verbal performance was maintained independently of balance challenge. The results of the first study provided statistical power estimates for the design of the second focussed experiment which made a within group (twenty four subjects) comparison of the impact of spatial versus verbal tasks on balancing on the hardest beam. The spatial task significantly elevated the incidence of falls whereas the verbal task had no effect on fall rate. The spatial task raised the incidence of falling by 50% (P = 0.0008) in comparison with 'no task'. The verbal task had no effect (P = 0.07). We conclude that sway magnitude is a poor index of multi-task load. Multi-tasking can increase the chance of falling and spatial processing may have a specific impact on balance. The significant elevation of fall frequency during cognitive tasking shows that the 'posture first' principal can be transgressed although the necessary condition for transgression may be that the subject is willing to take risks believing that he can arrest any fall.