DATASET VARIABLE DESCRIPTION DATA TYPE VALUE CODES

CANTAB sid unique subject identifier character

CANTAB age age of a subject at the time in clinic evaluation numeric N/A
CANTAB sex sex character F: female; M: male
There are nine stages to be completed in this task in the clinical setup. Subjects completing all
stages are deemed to have 'passed the test'. There are two key stages, the intra-dimensional
shift (stage 6) and the extra-dimensional shift (stage 8). Analysis of stage reached has often
been conducted using the likelihood ratio method for contingency tables which yields a
likelihood ratio statistic '2i' (for further details of this analysis see Robbins, T. 1977 in Iversen,
LL. et al. (Eds.) The Handbook of Psychopharmacology. Vol. 7, pp. 37-82, Plenum Press New
CANTAB IED Stages completed York). numeric N/A

This is a measure of the subject's efficiency in attempting the test. Thus, whilst a subject may
pass all nine stages, a substantial number of errors may be made in doing so. It is crucial to
note that subjects failing at any stage of the test by definition have had less opportunity to
CANTAB IED Total errors make errors - the IED Total errors (adjusted) measure attempts to compensate for this. numeric N/A

This is a measure of the subject's efficiency in attempting the test. Thus, whilst a subject may
pass all nine stages, a substantial number of errors may be made in doing so. It is crucial to
note that subjects failing at any stage of the test by definition have had less opportunity to
make errors. Therefore, this adjusted score is calculated by adding 25 for each stage not
attempted due to failure. This value of 25 is used since subjects must complete 50 trials to fail
CANTAB IED Total errors (adjusted) a stage and half of these could be correct by chance alone. numeric N/A
CANTAB PRM Percent correct This is the number of correct responses, expressed as a percentage. numeric N/A

This is errors made on stages sucessfully completed. Subjects failing at any stage of the ID/ED
shift will have less opportunity to make errors than those who finish, or get closer to finishing,
the task. It would therefore be misleading to simply compare errors made in circumstances
CANTAB IED Completed stage errors where different stages were reached (see IED total errors (adjusted)). numeric N/A

CANTAB IED Completed stage trials This is the number of trials on all successfully completed stages. numeric N/A
 Errors made in the extra-dimensional stage of the task are labelled EDS errors as they have
been committed at the stage where the subjects is required to make an extra-dimensional
CANTAB IED EDS errors shift. Errors committed at the reversal stage following the EDS stage are not included. numeric N/A
This metric records the number of errors made prior to the extra-dimensional shift of the
task. Errors are defined as instances when the subject fails to select the stimulus that is
CANTAB IED Pre-ED errors compatible with the current rule. numeric N/A
This is the number of trials completed on all attempted stages. Note that subjects failing at
any stage of the test have had less opportunity to complete trials - the IED Total trial (adjusted)
CANTAB IED Total trials measure attempts to compensate for this. numeric N/A
This is the number of trials completed on all attempted stages. The adjustment adds 50 for
CANTAB IED Total trials (adjusted) each stage not attempted due to failure at an earlier stage. numeric N/A

CANTAB PRM Mean correct latency This is the mean time to respond correctly. Latency is scored in milliseconds. numeric N/A

CANTAB PRM Number correct This is the number of correct responses (out of a maximum of 24 in the clinical setup). numeric N/A
This is the time taken to touch the stimulus after the press pad has been released and in trials
wherestimuli appear in one location only. Movement time latency is usually normally
CANTAB RTI Simple movement time distributed. numeric N/A

This is the speed with which the subject releases the press pad in response to the onset of a
stimulus in a single location. Reaction time latency is measured in milliseconds and tends
toward a positive skew. Five-choice reaction time latencies are reliably observed to be longer
than in simple reaction time. It should be remembered that subjects engaged in reaction time
tasks have the opportunity to make a variety of errors. Most are errors of commission ('too
soon', 'inaccurate' and 'wrong circle'), but it is possible to make an error of an omission by not
responding ('too late'). Latency tasks that contain accuracy demands require a trade-off
between speed and accuracy and so analysis of RT tasks need to consider making reference to
CANTAB RTI Simple reaction time both speed and accuracy. numeric N/A
B double prime is the signal detection measure of the strength of trace required to elicit a
response (range -1.00 to +1.00). Thus, it is the tendency to repond regardless of whether the
target sequence is present and uses the p(hit) and p(fa) results. A score close to +1.00
CANTAB RVP B'' indicates that the subject gave few false alarms. numeric N/A
 This measure details the mean time taken to respond and is reported in milliseconds. It only
CANTAB RVP Mean latency includes correct responses made within the response window of 1800 milliseconds. numeric N/A

The data from vigilance tasks such as RVP lend themselves to analysis using the principles of
Signal Detection Theory (SDT). The standard methodology of SDT allows for the analysis of the
two main components of the decision making process, the acquisition of information and the
criterion required for a response. In the case of RVP, the optimal pattern of response is to
maximise sensitivity so that so that no targets are missed and such that no false alarms are
committed. Clearly speed of response is a factor as subjects placing a premium on speed
rather than accuracy will tend to be less accurate in their detection of targets and their
correction rejections. The probability of a false alarm i.e. the subject responding
inappropriately, is equal to the total number of false alarms/(total false alarms + total correct
rejections)*100. A number of texts on the topic of SDT are available. Interested readers may
RVP Probability of false wish to browse Swets, J.A. (1996) 'Signal detection theory and ROC analysis in psychology and
CANTAB alarm diagnostics.' LEA numeric N/A

The data from vigilance tasks such as RVP lend themselves to analysis using the principles of
Signal Detection Theory (SDT). The standard methodology of SDT allows for the analysis of the
two main components of the decision making process, the acquisition of information and the
criterion required for a response. In the case of RVP, the optimal pattern of response is to
maximise sensitivity so that so that no targets are missed and such that no false alarms are
committed. Clearly speed of response is a factor as subjects placing a premium on speed
rather than accuracy will tend to be lees accurate in their detection of targets and their
correction rejections. The probability of a 'hit' i.e. the subject responding correctly, is
calculated from hits/(hits+misses)*100. A number of texts on the topic of SDT are available.
Interested readers may wish to browse Swets, J.A. (1996) 'Signal detection theory and ROC
CANTAB RVP Probability of hit analysis in psychology and diagnostics.' LEA numeric N/A
This is the number of stimuli that were correctly rejected, i.e. the number of stimuli that were
not part of a target sequence and were not responded to. This score is calculated from blocks
CANTAB RVP Total correct rejections 2, 3 and 4 only numeric N/A
 This records the number of times the subject respond outside the response window of a
CANTAB RVP Total false alarms target sequence This score is calculated from blocks 2, 3 and 4 only numeric N/A
This score represents the number of occasions upon which the target sequence is correctly
responded to (within a response window of 1800 milliseconds for the clinical setup). This
CANTAB RVP Total hits score is calculated from blocks 2, 3 and 4 only. numeric N/A

This score reports the number of occasions the subject fails to respond to a target sequence
CANTAB RVP Total misses within the response window. This score is calculated from blocks 2, 3 and 4 only numeric N/A

Subjects are encouraged to plan their moves before actually enacting the solution to the
problems Initial thinking time is the difference in time taken to select the first ball for the same
problem under the copy and follow conditions. Therefore, this gives an indication of the time
taken to plan the problem solution. This metric reports average initial thinking time for 2-move
problems. Please note that initial thinking time and subsequent thinking time may well
interact both with one another and with other SOC metrics, such as Minimum Move Solutions.
For example, some impulsive subjects may record very brief initial thinking time latencies, but
SOC Mean initial thinking fail to solve any of the presented problems. This score may be 0 if the subject is slower in the
CANTAB time (2 moves) 'follow' condition. numeric N/A

Subjects are encouraged to plan their moves before actually enacting the solution to the
problems. Initial thinking time is the difference in time taken to select the first ball for the
same problem under the copy and follow conditions. Therrefore, this gives an indication of the
time taken to plan the problem solutionThis metric reports average initial thinking time for 3-
move problems. Please note that initial thinking time and subsequent thinking time may well
interact both with one another and with other SOC metrics, such as Minimum Move Solutions.
For example, some impulsive subjects may record very brief initial thinking time latencies, but
SOC Mean initial thinking fail to solve any of the presented problems. This score may be 0 if the subject is slower in the
CANTAB time (3 moves) 'follow' condition. numeric N/A
 Subjects are encouraged to plan their moves before actually enacting the solution to the
problems. Initial thinking time is the difference in time taken to select the first ball for the
same problem under the copy and follow conditions. Therrefore, this gives an indication of the
time taken to plan the problem solutionThis metric reports average initial thinking time for 4-
move problems. Please note that initial thinking time and subsequent thinking time may well
interact both with one another and with other SOC metrics, such as Minimum Move Solutions.
For example, some impulsive subjects may record very brief initial thinking time latencies, but
SOC Mean initial thinking fail to solve any of the presented problems. This score may be 0 if the subject is slower in the
CANTAB time (4 moves) 'follow' condition. numeric N/A
This metric describes the mean number of moves required by the subject to solve problems
where the solution can be reached in a minimum of 2 moves (out of a possible two 2-move
CANTAB SOC Mean moves (2 moves) problems in the clinical setup). numeric N/A
This metric describes the mean number of moves required by the subject in problems where
the solution can be reached in a minimum of 3 moves (out of a possible 2 3-move problems in
CANTAB SOC Mean moves (3 moves) the clinical setup). numeric N/A
This metric describes the mean number of moves required by the subject in problems where
the optimal solution can be reached in a minimum of 4 moves (out of a possible four 4-move
CANTAB SOC Mean moves (4 moves) problems in the clinical setup). numeric N/A
This metric describes the mean number of moves required by the subject in problems where
the optimal solution can be reached in a minimum of 5 moves (out of a possible four 5-move
CANTAB SOC Mean moves (5 moves) problems in the clinical setup). numeric N/A

This measure reflects the subject's speed of movement after the initial move has been made
for two move problems.. This metric is obtained by calculating the difference in time between
selecting the first ball and completing the problem for the same problem under the two
conditions (copy and follow) and dividing this result by the number of moves [Link] note
that initial thinking time and subsequent thinking time may well interact both with one
another and with other SOC metrics, such as Minimum Move Solutions. For example, some
SOC Mean subsequent impulsive subjects may record very brief initial thinking time latencies, but fail to solve any of
CANTAB thinking time (2 moves) the presented problems. This score may be 0 if the subject is slower in the 'follow' condition. numeric N/A
 This measure reflects the subject's speed of movement after the initial move has been made
for three move problems. This metric is obtained by calculating the difference in time between
selecting the first ball and completing the problem for the same problem under the two
conditions (copy and follow) and dividing this result by the number of moves [Link] note
that initial thinking time and subsequent thinking time may well interact both with one
another and with other SOC metrics, such as Minimum Move Solutions. For example, some
SOC Mean subsequent impulsive subjects may record very brief initial thinking time latencies, but fail to solve any of
CANTAB thinking time (3 moves) the presented problems. This score may be 0 if the subject is slower in the 'follow' condition. numeric N/A

This measure reflects the subject's speed of movement after the initial move has been made
for four move problems. This metric is obtained by calculating the difference in time between
selecting the first ball and completing the problem for the same problem under the two
conditions (copy and follow) and dividing this result by the number of moves [Link] note
that initial thinking time and subsequent thinking time may well interact both with one
another and with other SOC metrics, such as Minimum Move Solutions. For example, some
SOC Mean subsequent impulsive subjects may record very brief initial thinking time latencies, but fail to solve any of
CANTAB thinking time (4 moves) the presented problems. This score may be 0 if the subject is slower in the 'follow' condition. numeric N/A

CANTAB SRM Mean correct latency This is the mean time to respond correctly. Latency is scored in milliseconds. numeric N/A
CANTAB SRM Number correct This is the number of correct responses (out of a possible 20 in the clinical setup). numeric N/A

SWM Between errors (4 Between errors are defined as occasions upon which the subject revisits a box in which a
CANTAB boxes) token has previously been found. This metric records between errors for 4 box problems only. numeric N/A

SWM Between errors (6 Between errors are defined as occasions upon which the subject revisits a box in which a
CANTAB boxes) token has previously been found. This metric records between errors for 6 box problems only. numeric N/A

SWM Between errors (8 Between errors are defined as occasions upon which the subject revisits a box in which a
CANTAB boxes) token has previously been found. This metric records between errors for 8 box problems only. numeric N/A
 These are occasions where the subject has committed an error that can be categorised as
CANTAB SWM Double errors both a within and a between error, This is calculated for trials of four or more tokens only. numeric N/A

SWM Double errors (4 These are occasions where the subject has committed an error that can be categorised as
CANTAB boxes) both a within and a between error, summarised for 4 box problems. numeric N/A

SWM Double errors (6 These are occasions where the subject has committed an error that can be categorised as
CANTAB boxes) both a within and a between error, summarised for 6 box problems. numeric N/A

SWM Double errors (8 These are occasions where the subject has committed an error that can be categorised as
CANTAB boxes) both a within and a between error, summarised for 8 box problems. numeric N/A
This is the number of times a box is selected that is certain not to contain a blue token and
therefore should not have been visited by the subject, i.e. between errors + within errors -
CANTAB SWM Total errors double errors. numeric N/A
Within errors are defined as the number of errors made within a search, [Link] number of
times a subject revisits a box already found to be empty during the same search. . This is
CANTAB SWM Within errors calculated for trials of four or more tokens only. numeric N/A

Within errors are defined as the number of errors made within a search, i.e. repeated
SWM Within errors (4 responses to a box previously opened and shown to be empty earlier in the same search
CANTAB boxes) sequence. This metric records within errors made at in 4-box trials. numeric N/A

Within errors are defined as the number of errors made within a search, i.e. repeated
SWM Within errors (6 responses to a box previously opened and shown to be empty earlier in the same search
CANTAB boxes) sequence. This metric records within errors made at in 6-box trials. numeric N/A

Within errors are defined as the number of errors made within a search, i.e. repeated
SWM Within errors (8 responses to a box previously opened and shown to be empty earlier in the same search
CANTAB boxes) sequence. This metric records within errors made at in 8-box trials. numeric N/A
CANTAB SRM Percent correct This is the number of correct responses, expressed as a percentage. numeric N/A
 This is the time taken to touch the stimulus after the press pad has been released and in trials
RTI Five-choice movement where one of five possible different stimuli have been presented. Movement time latency is
CANTAB time usually normally distributed for correct responses. numeric N/A

This is the speed with which the subject releases the press pad in response to a stimulus in any
one of five locations. Choice reaction time latency is measured in milliseconds and tends
toward a positive skew. Five-choice reaction time latencies are reliably observed to be longer
than in simple reaction time. It should be remembered that subjects engaged in reaction time
tasks have the opportunity to make a variety of errors. Most are errors of commission ('too
soon', 'inaccurate' and 'wrong circle'), but it is possible to make an error of an omission by not
responding ('too late'). Latency tasks that contain accuracy demands require a trade-off
RTI Five-choice reaction between speed and accuracy and so analysis of RT tasks need to consider making reference to
CANTAB time both speed and accuracy. numeric N/A
This is the signal detection measure of sensitivity to errors, regardless of error tendency
(range 0.00 to 1.00; bad to good). In essence, this metric is a measure of how good the subject
CANTAB RVP A' is at detecting target sequences using p(hit) and p(fa). numeric N/A

Subjects are encouraged to plan their moves before actually enacting the solution to the
problems Initial thinking time is the difference in time taken to select the first ball for the same
problem under the copy and follow conditions. Therrefore, this gives an indication of the time
taken to plan the problem solutionThis metric reports average initial thinking time for 5-move
problems. Please note that initial thinking time and subsequent thinking time may well interact
both with one another and with other SOC metrics, such as Minimum Move Solutions. For
example, some impulsive subjects may record very brief initial thinking time latencies, but fail
SOC Mean initial thinking to solve any of the presented problems. This score may be 0 if the subject is slower in the
CANTAB time (5 moves) 'follow' condition. numeric N/A
 This measure reflects the subject's speed of movement after the initial move has been made
for five move problems. This metric is obtained by calculating the difference in time between
selecting the first ball and completing the problem for the same problem under the two
conditions (copy and follow) and dividing this result by the number of moves [Link] note
that initial thinking time and subsequent thinking time may well interact both with one
another and with other SOC metrics, such as Minimum Move Solutions. For example, some
SOC Mean subsequent impulsive subjects may record very brief initial thinking time latencies, but fail to solve any of
CANTAB thinking time (5 moves) the presented problems. This score may be 0 if the subject is slower in the 'follow' condition. numeric N/A

This is a fundamental metric test, recording the number of occasions upon which the subject
has successfully completed a test problem in the minimum possible number of moves. For the
clinical setup, this is scored out of a possible 12 problems since eight practice problems are
SOC Problems solved in excluded from the calculation (the first six problems in the first block and the first two
CANTAB minimum moves problems in the second block). numeric N/A

Between errors are defined as times the subject revisits a box in which a token has previously
CANTAB SWM Between errors been found. This is calculated for trials of four or more tokens only. numeric N/A

Owen et al. (Neuropsychologia 1990:28; 1021-1034) have suggested that an efficient strategy
for completing this task is to follow a predetermined sequence by beginning with specific box
and then, once a blue token has been found, to return to that box to start the new search
sequence. An estimate of the use of this strategy is obtained by counting the number of times
the subject begins a new search with the same box. A high score represents poor use of this
strategy and a low score equates to effective use. Thus, for the clinical setup, the minimum
CANTAB SWM Strategy strategy score is 1 for each stage (i.e.8) and the maximum is 1 for each search (i.e. 56). numeric N/A
1: IED test failed; 2:
RVP test failed; 3: IED
CANTAB Warnings warning codes numeric test aborted