Imaging the mental components of a planning task.
Rowe JB., Owen AM., Johnsrude IS., Passingham RE.
The Tower of London task (TOL) has been widely used to assess the ability to plan. We used H(2)O(15)-positron emission tomography to isolate some of the cognitive components of the task. Ten male volunteers were scanned twice in each of six conditions. In two conditions (plan) the subjects had to plan the best solution to TOL problems. In two other conditions (plan-control) the subjects were required to generate four moves without being constrained by a goal. In plan and plan-control tasks the subjects either planned the moves and then executed them (MOVE conditions) or imagined the necessary moves (IMAGINE conditions). The plan and plan-control tasks were matched for the working memory load and "initial thinking time". A visuomotor control task and rest served as baseline conditions. Performance on the plan tasks, in contrast to the baseline conditions, was associated with activation in the dorsal prefrontal cortex, premotor and parietal cortex, and cerebellum. Performance of the plan-control tasks was associated with activation of the same areas. Contrasting the plan with the plan-control tasks revealed no residual activation in the prefrontal cortex. These data show that the activity of the dorsolateral prefrontal cortex on the TOL can be accounted for by the components of generating, selecting and/or remembering mental moves. The task of relating the moves to the goal involves a comparison with a representation of the goal in posterior association areas. We did not find evidence that activation of the dorsal prefrontal cortex is specifically related to the evaluation of a path towards a specified goal, a key component of planning.