Abstract
Although the Tower of London (TOL) has been extensively used to assess planning ability in patients with Parkinson's disease (PD), the reported presence or extent of any planning deficits has been inconsistent. This may partly be due to the heterogeneity of the TOL tasks used and a failure to consider how structural problem parameters may affect task complexity. In the present study, planning in PD patients was assessed by systematically manipulating TOL problem structure. Results clearly disprove the identity assumption of problems with an equal number of minimum moves. Instead, substantial parts of planning performance were related to more subtle aspects of problem structure, such as subgoaling patterns and goal hierarchy. Planning in PD patients was not impaired in general but was affected when the information provided by the problem states was ambiguous in terms of the sequential order of subgoals, but not by increases in search depth.