Sustained Attention to Response Test (SART) - anyone knowing how to analyze the data?


Sustained Attention to Response Test (SART) - anyone knowing how to...
Author
Message
Christina Bengtsson
Christina Bengtsson
Partner Member (929 reputation)Partner Member (929 reputation)Partner Member (929 reputation)Partner Member (929 reputation)Partner Member (929 reputation)Partner Member (929 reputation)Partner Member (929 reputation)Partner Member (929 reputation)Partner Member (929 reputation)
Group: Forum Members
Posts: 8, Visits: 38
Hi, 
Do you happen to know how or from where I can get an explanation tool or likewise for how to analyse the different outcome (data) of the Sustained Attention to Response Test (SART)? 
Attachments
Dave
Dave
Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)
Group: Administrators
Posts: 13K, Visits: 104K
You'll find detailed explanations of the script's data output in the comments inside the script (starting at line 269 and excerpted below):

*******************************************************************************************************************
*******************************************************************************************************************
    DATA: this section contains data file information
*******************************************************************************************************************
*******************************************************************************************************************

********************
raw data
********************
date, time, subject, group:            date and time script was run with the current subject/groupnumber
blockcode, blocknum:                the name and number of the current block
trialcode, trialnum:                 the name and number of the currently recorded trial
                                        (Note: not all trials that are run might record data)
/digitpresentationtime:                the amount of time the digits are presented (default: 250ms)
/maskpresentationtime:                the amount of time the mask is presented (default: 900ms)

/trialtype:                            "Go" (digit !=3); "NoGo" (digit == 3)
/digit:                                contains the currently selected digit
/fontsize:                            contains the currently (randomly) selected fontsize
response:                            the participant's response
                                        SART trials: 0 for no response or 57 for Spacebar
                                        Probe trials (if run): the selected anchor digit
correct:                            the correctness of the response
RT:                                    the latency of the response (= trial latency unless no response was given; no response leaves values.RT empty)
latency:                             the trial latency in ms (if no response, trial latency shows the trialduration)
/latencytype:                        0 = no response given (suppression); 1 = anticipatory latency (< values.anticipatoryresponsetime);
                                    2 = ambiguous latency; 3 = valid latency  (>= values.validresponsetime)
                                    Note: independent of Go-NoGo trialtype
/responsetype:                        "NoGo Success": correctly suppressed response to NoGo trial (digit 3)
                                    "Omission": incorrectly suppressed response to a Go trial (digit other than 3)
                                    "NoGo Failure": any response to a NoGo trial
                                    "Go Anticipatory": anticipatory response for Go trials with latencies < values.anticipatoryresponsetime
                                    "Go Ambiguous": ambiguous response for Go trials
                                    "Go Success": valid response for Go trials with latencies >= values.validresponsetime
/count_anticipatory:                counts the number of times a Go-latency was faster than values.anticipatoryresponsetime
/correctsuppressions:                counts number of correct suppressions (-> no response to digit 3)
/incorrectsuppressions:                counts number of incorrect suppressions (-> no response to digit other than 3)
/count_NoGo:                        counts NoGo trials (digit 3)
/count_Go:                            counts Go trials (digits other than 3)
/count_validGo:                        counts Go trials with a correct response and latencies >= values.validresponsetime
/countprobes:                        counts the number of probes run

radiobuttons.difficulty.response:
radiobuttons.interest.response:            responses to the posttask survey questions

...

********************
summary data
********************
script.startdate:                date script was run
script.starttime:                time script was started
script.subjectid:                subject id number
script.groupid:                    group id number
script.elapsedtime:                time it took to run script (in ms)
/completed:                        0 = script was not completed (prematurely aborted); 1 = script was completed (all conditions run)

radiobuttons.difficulty.response:
radiobuttons.interest.response:            responses to the posttask survey questions

/percent_NoGosuccess:                    percentage of correct suppressions in NoGo trials (=NoGo success)
/percent_omissions:                        percentage of incorrect suppressions in Go trials (=omission of response)

Mean RTs are calculated for the correct consecutive (not interrupted either by a NoGo or by an omission trial) last four Go- trials (digit other than 3) preceding successful NoGo trial
Mean RTs are calculated for the correct consecutive (not interrupted either by a NoGo or by an omission trial) last four Go- trials (digit other than 3)  preceding preceding failed NoGo trial
!!!!Note: in this script any 4 correct Go trials are counted irrespective of values.latencytype

/meanRT_GObeforesuccessNOGO:                mean latency of consecutive 4 correct trials before correct suppression of response to digit 3
                                            (=> a measure of speed before successful NoGo trials)
/meanRT_GObeforefailedNOGO:                    mean latency of consecutive 4 correct trials before incorrect response of response to digit 3
                                            (=> a measure of speed before failed NoGo trials)

/meanRT_go:                                    mean latency of valid and correct Go trials (latencies >= values.validresponsetime)
/stdRT_go:                                    estimated standard deviation (STD) of valid and correct Go trials
/CV_go:                                        coefficient of variablity (CV = STD/Mean) => a measure of variability independent of mean differences
GO

Merge Selected

Merge into selected topic...



Merge into merge target...



Merge into a specific topic ID...




Reading This Topic

Explore
Messages
Mentions
Search