Is random tracking more difficult in two dimensions than one dimension?

There is conflicting evidence in the literature as to whether sensory-motor performance degrades on going from a 1D to a 2D task. We undertook a study to investigate more closely the fundamental questions: Is random tracking more difficult in 2D than in 1D? If so, is the increased difficulty due to the increased velocity or the increased dimensionality of the 2D task? The study involved 16 normal subjects tracking 2D random targets and two orthogonal 1D equivalents using a floor-mounted joystick. To investigate the speed versus dimensionality question, a version of the standard 2D task was devised which used a scaled target signal with the same average velocity and displacement as the 1D task. Tracking performance (mean absolute error) was found to degrade by 41% (p<.001) on going from the 1D to the scaled 2D task, This demonstrates that tracking is indeed more demanding in 2D than 1D and that this can be unequivocally attributed to the increased dimensionality of the 2D task. Most of the poorer performance on the 2D task could be attributed to an extra 59 ms in tracking delay and is considered most likely due to the extra degree of freedom reducing the ability of the sensory-motor system to offset inherent delays in the system by using prediction.