METHODS: In experiment 1 (n = 10), we tested the direction of force exerted in an isometric aiming task before and after 40 repetitions of 2-s maximal-force ballistic contractions toward a single directional target. In experiment 2 (n = 12), each participant completed three training conditions in a counterbalanced crossover design. In two conditions, both the aiming task and the training were conducted in the same (neutral) forearm posture. In one of these conditions, the training involved weak forces to determine whether the level of neural drive during training influences the degree of bias. In the third condition, high-force training contractions were performed in a 90° pronated forearm posture, whereas the low-force aiming task was performed in a neutral forearm posture. This dissociated the extrinsic training direction from the pulling direction of the trained muscles during the aiming task.

RESULTS: In experiment 1, we found that aiming direction was biased toward the training direction across a large area of the work space (approximately ±135°; tested for 16 targets spaced 22.5° apart), whereas in experiment 2, we found systematic bias in aiming toward the training direction defined in extrinsic space, but only immediately after high-force contractions.