Random duration of trial


Author
Message
***JayD***
***JayD***
Distinguished Member (4K reputation)Distinguished Member (4K reputation)Distinguished Member (4K reputation)Distinguished Member (4K reputation)Distinguished Member (4K reputation)Distinguished Member (4K reputation)Distinguished Member (4K reputation)Distinguished Member (4K reputation)Distinguished Member (4K reputation)
Group: Forum Members
Posts: 17, Visits: 67
Hi,

I have a couple of questions

1) Is it possible to randomize a trial's duration (either using /trialduration or /response = timeout ())? I want to try and have my intertrial interval randomly vary between 500ms and 400ms on any given trial.

2) I want a particular (mouse-click) response to give participants a 'reward' feedback trial on 80% of trials, while presenting the same mouse-click with a different 'no-reward' feedback trial on the other 20%. I'm not sure how to go about building in these percentages into the branch trials. What's the best way to do this?

Thanks,

Jay
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: 98K
Re. #1: Yes. You can do something like

<values>
/ randomiti = 0
...
</values>

with

<trial iti>
/ ontrialbegin = [values.randomiti = round(rand(400,500))]
...
/ trialduration = values.randomiti
...
</trial>

/pre- and /posttrialpause, /timeout and /trialduration all accept values which you can set at will via /ontrialbegin etc.

For the relationship between the various timing components involved in a <trial> see the "How to Control Trial Duration and Inter-Trial Intervals" topic in the documentation.

Re. #2: Not sure why you would need any elaborate /branch logic, but the question is underspecified. For example, it's not clear if the "specific response" is the only allowed response; if not, you cannot possibly know how many times a participant will actually provide the "specific response", i.e., it may not even be possible to achieve the 80/20 split. Suppose the participant only provides the "specific" response twice across however many trials you have: You could at best achieve a 50/50 split in that situation.

If, otoh, there is only that single "specific response", you can simply do

<block someblock>
/ trials = [1-10 = noreplace(rewardtrial, rewardtrial, rewardtrial, rewardtrial, norewardtrial)]
...
</block>

which will run 8 "reward" trials and 2 "no reward" trials. No branching required.



Edited 8 Years Ago by Dave
GO

Merge Selected

Merge into selected topic...



Merge into merge target...



Merge into a specific topic ID...




Reading This Topic

Explore
Messages
Mentions
Search