The cost of shifting tasks across consecutive trials was examined as a func
tion of the time interval between a task-cue and a subsequent target (cue-t
o-target interval, CTI). Task-shift costs are assumed to index a process by
which subjects reconfigure their task-set when they must perform different
tasks across consecutive trials. In neurologically intact teenagers, incre
ased CTI causes reduced task-shift costs since subjects have more time to r
econfigure their task-set prior to target onset. Some closed head injury ad
olescents however, show no reduction of task-shift costs with increased CTI
. This indicates that subjects are incapable of reconfiguring their task-se
t from the cue alone, they also require the target before they are able to
initiate this reconfiguration. (C) 2000 Academic Press.