The impact of basal ganglia dysfunction on cognitive skill learning was explored using a learning version of the
Tower of London (TOL) task, which places a heavy load on working memory and is not confounded by declarative memory,
as have been previous tasks. Two subgroups of Parkinson’s disease (PD) patients were assessed and also completed a
selection of neuropsychological tests: the first was unmedicated (de novo, n=12) and the second included patients normally
receiving L-DOPA, but tested off medication (n=12). Overall, neither subgroup was impaired when learning the
task compared to control participants (n=22). Six patients, however, failed to improve their performance with practice.
Their learning deficit could not be explained in terms of their functional status; instead, it was related to deficits on span
tests. Thus, the inability to acquire a new cognitive skill in PD may not be due to learning impairments per se, but rather,
it appears to be secondary to working memory deficits.