It has long been argued that patients with lesions in the prefrontal c
ortex have difficulties in decision making and problem solving in real
-world, ill-structured situations, particularly problem types involvin
g planning and lookahead components. Recently, several researchers hav
e questioned our ability to capture and characterize these deficits ad
equately using just the standard neuropsychological test batteries, an
d have called for tests that reflect real-world task requirements more
accurately. We present data from 10 patients with focal lesions to th
e prefrontal cortex and 10 normal control subjects engaged in a real-w
orld financial planning task. We also introduce a theoretical framewor
k and methodology developed in the cognitive science literature for qu
antifying and analysing the complex data generated by problem-solving
tasks. Our findings indicate that patient performance is impoverished
at a global level but not at the local level. Patients have difficulty
in organizing and structuring their problem space. Once they begin pr
oblem solving, they have difficulty in allocating adequate effort to e
ach problem-solving phase. Patients also have difficulty dealing with
the fact that there are no right or wrong answers nor official termina
tion points in real-world planning problems. They also find it problem
atic to generate their own feedback. They invariably terminate the ses
sion before the details are fleshed out and all the goals satisfied Fi
nally, patients do not take full advantage of the fact that constraint
s on real-world problems are negotiable. However it is not necessary t
o postulate a 'planning' deficit It is possible to understand the pati
ents' difficulties in real world planning tasks in terms of the follow
ing four accepted deficits: inadequate access to 'structured event com
plexes', difficulty in generalizing from particulars, failure to shift
between 'mental sets', and poor judgment regarding adequacy and compl
eteness of a plan.