The authors examined the effects of induced conceptions of ability on motor learning. Participants in 3 groups practiced a balance task after receiving instructions suggesting that the task would reflect an inherent ability (IA group), represent an acquirable skill (AS group), or no ability-related instructions (control group). Across 2 days of practice, the AS and IA groups showed greater improvement in performance compared with the control group. For the retention test on Day 3, the AS group tended to demonstrate generally more effective balance performance than the control group and increasingly greater effectiveness compared with the IA group. Moreover, AS group participants made higher-frequency (reflexive) movement adjustments than participants of the other 2 groups, indicating a greater automaticity in the control of their movements. Thus, learning was enhanced by instructions portraying the task as a learnable skill, rather than revealing a fixed inherent capacity or no instructions (control group).