Stroop minlatency


Author
Message
Nicola
Nicola
Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)
Group: Forum Members
Posts: 11, Visits: 1

Hello all,


I'm trying to modify the stroop script so that I can easily identify outliers. So, I've introduced a minimum latency and maximum latency variable for both incongruent and control trials. As far as I can see, the maximum latency is working fine. The minimum latency, however, does not update correctly after each trial. Sometimes it updates with a latency that is larger and sometimes it doesn't update when there is a smaller latency. Any thoughts on this? Is there more to the minlatency function that I'm not aware of?


Also, would this be affecting my variance/SD function? I've included these so that Inquisit will alert me when there are latencies < 300ms or > mean + 3SD.


Thank you!



<expressions>
...
/ minincongruentlatency = minlatency(trial.blueincongruent, trial.redincongruent, trial.greenincongruent, trial.blackincongruent)
/ maxincongruentlatency = maxlatency(trial.blueincongruent, trial.redincongruent, trial.greenincongruent, trial.blackincongruent)
/ varincongruentlatency = varlatency(trial.blueincongruent, trial.redincongruent, trial.greenincongruent, trial.blackincongruent)
/ sdincongruentlatency = sqrt(expressions.varincongruentlatency)
/ cutoffincongruentlatency = expressions.incongruentlatency+(3*expressions.sdincongruentlatency)/ mincontrollatency = minlatency(trial.bluecontrol, trial.redcontrol, trial.greencontrol, trial.blackcontrol)
/ maxcontrollatency = maxlatency(trial.bluecontrol, trial.redcontrol, trial.greencontrol, trial.blackcontrol)
/ varcontrollatency = varlatency(trial.bluecontrol, trial.redcontrol, trial.greencontrol, trial.blackcontrol)
/ sdcontrollatency = sqrt(expressions.varcontrollatency)
/ cutoffcontrollatency = expressions.controllatency+(3*expressions.sdcontrollatency)
...
</expressions>


Tags
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: 12K, Visits: 83K

As far as I am aware (and according to the functions reference in the documentation):


minlatency() = minimum response latency in the current block


vs.


totalminlatency() = minimum response latency (i.e. across all blocks)


Same for maxlatency vs. totalmaxlatency, meanlatency vs. totalmeanlatency etc.


Other than that, there's not much I could say just by staring at the few lines you posted / without any actual data. If you want full control over metrics and do not want to rely on built-in functions, I would recommend you implement min and max as needed by using values and expressions.


Nicola
Nicola
Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)
Group: Forum Members
Posts: 11, Visits: 1

Hi Dave,


Thank you for your reply. I only have two blocks - practice and test. So, I assume it's not an issue between the blocks. Plus, it seems that maxlatency is working but not minlatency. It seems that minlatency is updated (correctly or incorrectly) on the first 'blackincongruent' and 'blackcontrol' trials. I don't know what else it is doing the rest of the time (nor why it doesn't update on the first instances of red, green or blue). I have attached data from two attempts at the experiment. If you have time to have a look, that would be great!


I may have to use values/expressions. I think I can do this for minlatency and maxlatency, but I'm not sure about variance (and standard deviation). Do you have any further information regarding how the variance is calculated? Is it taking the sum of squared differences between each trial latency and the final mean for the specified trials? Or is it a sum of diffferences from the mean at the time of that trial? If the latter, it's probably not of use in this case anyway.


Thank you!


Attachments
Stroop data.xlsx (587 views, 76.00 KB)
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: 12K, Visits: 83K

but I'm not sure about variance (and standard deviation). Do you have any further information regarding how the variance is calculated?


See e.g how SD is calculated in any of the IAT scripts.


Or is it a sum of diffferences from the mean at the time of that trial?


Well, it must be, regardless of how you calculate it (built-in function vs expression). The final mean isn't (and cannot be) known until the experiment is over and all data are in.


Nicola
Nicola
Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)Esteemed Member (1.8K reputation)
Group: Forum Members
Posts: 11, Visits: 1

Hi Dave,


Thank you for all your help. I have decided to compute the SD at the end, using SPSS.


Nicola


GO

Merge Selected

Merge into selected topic...



Merge into merge target...



Merge into a specific topic ID...




Reading This Topic

Explore
Messages
Mentions
Search