738 T.SHALLICE AND P.W.BURGESS of unforeseen subgoals and hence plan modification;this would also be necessary if unforeseen events occur.These stages are illustrated in the simple scenario shown in Table 6. The present 2 multiple-subgoal tests appear to require all these processes and many of the errors that occur in the execution of the tests appear to occur from a failure of one or other of them,although quite frequently it is not possible to be precise about which process is impaired.However,if one groups the above processes into three types then a provisional grouping of the lapses that occur in the 2 experimental tasks can be made,although possible overlaps between the three types still exist.The types are: (1)Plan Formulation or Modification;(2)Marker Creation or Triggering;(3)Evaluation and Goal Articulation (see Table 6 for an example).Table 7 gives examples from the TABLE7.EXAMPLE OF THE THREE微TEO HADE BY THE PATIENTS ON THE Plan formularionor mdifc Marker creation or triggering efore all relevant information ing out of bo t of ME (Case 3 in M E(Cae52,3) g 5 Evaluation and goal ariculation Thinkin E whe had o ncorrectly (Cases 2.3) SE Six Elements:ME Multiple Errands 2 experimental tasks of the three ty es of lapse.It can be seen that there are exan nle of all three types bei nade by each of the patients but the e prevale e of ule br on April 8.2016 marker errors rtan problem.it is also a parent that the marker errors can be subd vided.In s 1 the marker has to re nd to a t in some to a social co and in so me to a task rule cue (2.4).In all thes lead to the inhibition of the on marke should ur but in othe the should have led to a ur Ho r th ore ot with ntrol subi 、t ll E when something was bo aght in a shop (all 3 atie ts but also 5/9 co If the everyday life behaviour of the ned.it is often less eas categorize r any of the lap that they time Ne ertheless th Y anifest seem likely to arise from as their difficulties in the tal task The processe of plan and e valuation have often been ascribed to frontal738 T. SHALLICE AND P. W. BURGESS of unforeseen subgoals and hence plan modification; this would also be necessary if unforeseen events occur. These stages are illustrated in the simple scenario shown in Table 6. The present 2 multiple-subgoal tests appear to require all these processes and many of the errors that occur in the execution of the tests appear to occur from a failure of one or other of them, although quite frequently it is not possible to be precise about which process is impaired. However, if one groups the above processes into three types then a provisional grouping of the lapses that occur in the 2 experimental tasks can be made, although possible overlaps between the three types still exist. The types are: (1) Plan Formulation or Modification; (2) Marker Creation or Triggering; (3) Evaluation and Goal Articulation {see Table 6 for an example). Table 7 gives examples from the TABLE 7. EXAMPLE OF THE THREE TYPES OF LAPSE MADE BY THE PATIENTS ON THE EXPERIMENTAL TASKS Plan formulation or modification Marker creation or triggering 1. Going to Post Office before all relevant information 1. A very extended period spent on a single SE subtask obtained (Cases 1-3) without a break (Cases. 1,3) 2. No plan for SE (Cases 1 -3) 2. Going out of bounds in ME (Cases 2, 3) 3. Failure to look at watch at start of ME (Case 3) 3. Attempt not to pay for item in shop (Casel, 2) 4. Not knowing quite how he would get to X and there- 4. Looking in irrelevant shops (Cases 1, 2) fore not tackling SE. Journey back subtask (Case 1) 5. Generation of inappropriate criteria in an ME subtask 5. Not using the notes made for 4 min about journey at (Case 3) start of SE Evaluation and goal articulation 1. Posting postcard without all information (Cases 2, 3) 2. Failure to keep a check of money in ME (Cases 2, 3) 3. Thinking finished ME when had not (Case 1) 4. Ending SE attempt 1 min early when far from having completed subtasks (Case 3) 5. Carrying out SE subtasks incorrectly (Cases 2, 3) SE = Six Elements; ME = Multiple Errands. 2 experimental tasks of the three types of lapse. It can be seen that there are examples of all three types being made by each of the patients, but the prevalence of rule breaks in the behaviour of all 3 patients suggest that marker errors are a particularly important problem. It is also apparent that the marker errors can be subdivided. In some (1,5) the marker has to respond to a temporal cue, in some to a social convention cue (3) and in some to a task rule cue (2, 4). In all these cases triggering the marker should lead to the inhibition of the ongoing behaviour, but in other cases triggering the marker should have led to a positive behaviour. However, for this type of error there was much more overlap with control subjects (e.g., not coming to tell E when something was bought in a shop (all 3 patients, but also 5/9 controls)). If the everyday life behaviour of the patients is examined, it is often less easy to categorize many of the lapses that they make because they involve a larger period of time. Nevertheless, the problems they manifest seem likely to arise from the same source as their difficulties in the experimental task. The processes of plan formulation and evaluation have often been ascribed to frontal by guest on April 8, 2016 http://brain.oxfordjournals.org/ Downloaded from