No matter how hard you practise a movement, you can never be entirely sure how it will turn out. Shouldn't the same action executed under the same conditions always produce the same result? Yet even professional darts players, throwing in a controlled indoor environment and standing a set distance from the board, can miss the bull's-eye.

Credit: PHOTODISC/PHOTOLIBRARY

Many theories of muscle control have assumed that such errors arise from variation generated during the movement — particularly 'noise' in the way that neurons pass instructions to the muscles at the neuromuscular junction. But Mark Churchland, Afsheen Afshar and Krishna Shenoy report that a large part of the problem could instead arise as the brain plans the action (Neuron 52, 1085–1096; 2006).

They observed monkeys reaching for visual targets that appeared on a screen. When a target first appeared, it jittered slightly in place, and the animals were trained not to reach for it until it became stationary a half-second to a second later — allowing a period of preparation.

The authors recorded neural activity from the motor cortex and the premotor cortex, two brain regions involved in movement planning and execution. Comparing the monkeys' reaching movements with these recordings, they found that variations in the velocity of the reaches correlated with fluctuations in brain activity during the preparatory period — hundreds of milliseconds before the movement started.

So it seems that the execution of even a simple, well-practised task is limited by the brain's ability to plan the same movement over and over again. Indeed, Churchland and colleagues estimate that this constraint could account for at least half of the variability in the monkeys' movements.

Whether the fluctuations they observe actually arise in the premotor and motor cortex, or merely reflect variations elsewhere in the brain, is still an open question. And how variations in sensory input might affect the subsequent movement has yet to be fully explored. Finding the answers will have implications for our understanding of how the brain controls movement, and in the long term could have an impact on how movement disorders are treated.