Consolidation patterns of human motor memory.  Sarah E. Criscimagna-Hemminger and Reza Shadmehr (2008).  Journal of Neuroscience.        

Abstract  Can memories be unlearned, or is unlearning a form of acquiring a new memory that competes with the old, effectively masking it?  We considered motor memories that were acquired when people learned to use a novel tool.  We trained people to reach with tool A and quantified recall in error-clamp trials, i.e., trials in which the memory was re-activated but error-dependent learning was minimized.  We measured both the magnitude of the memory and its resistance to change.  With passage of time between acquisition and re-activation (up to 24 hrs), memory of A slowly declined, but upon re-activation remained resistant to change.  Following learning of A, brief exposure to B brought performance back to baseline, i.e., apparent extinction.  Yet, for up to a few minutes after A+B training, output in error-clamp trials increased from baseline to match those who had trained only in A.  This spontaneous recovery and convergence demonstrated that B did not produce any unlearning of A.   Rather, it masked A with a new memory that was very fragile.  We tracked the memory of B as a function of time and found that within minutes it was transformed from a fragile to a more stable state.  Therefore, a sudden performance error in a well-learned motor task does not produce unlearning, but rather installs a competing but fragile memory that with passage of time acquires stability.  Learning not only engages processes that adapt at multiple timescales, but that once practice ends, the fast states are partially transformed into slower states.