Pseudo-randomization


Author
Message
helenr
helenr
Esteemed Member (1.6K reputation)Esteemed Member (1.6K reputation)Esteemed Member (1.6K reputation)Esteemed Member (1.6K reputation)Esteemed Member (1.6K reputation)Esteemed Member (1.6K reputation)Esteemed Member (1.6K reputation)Esteemed Member (1.6K reputation)Esteemed Member (1.6K reputation)
Group: Awaiting Activation
Posts: 11, Visits: 54
Hi !

I was wondering how I can present the following trials so that trialtype'1' (for e.g. right100_win1) is not consecutively followed by another trialtype1 (so an acceptable sequence could be 'right100_win1, right100_win4, right50_win1')

Thanks for your help :-)


<block main_exp>
/trials = [1-60,62-121=noreplacenorepeat (
right100_win1,right100_win1, left100_win1,left100_win1,right50_win1,left50_win1, right50_lose1, left50_lose1, right0_lose1,right0_lose1,left0_lose1,left0_lose1,
right100_win2,right100_win2, left100_win2,left100_win2, right50_win2, left50_win2, right50_lose2, left50_lose2,right0_lose2,right0_lose2,left0_lose2,left0_lose2,
right100_win3,right100_win2, left100_win3,left100_win2, right50_win3, left50_win2, right50_lose3, left50_lose3,right0_lose3,right0_lose3,left0_lose3,left0_lose3,
right100_win4,right100_win4, left100_win4,left100_win4, right50_win4, left50_win4, right50_lose4, left50_lose4,right0_lose4,right0_lose4,left0_lose4,left0_lose4,
right100_win5,right100_win5, left100_win5,left100_win5, right50_win5, left50_win5, right50_lose5, left50_lose5,right0_lose5,right0_lose5,left0_lose5,left0_lose5,
right100_win6,right100_win6, left100_win2,left100_win6, right50_win6, left50_win6, right50_lose6, left50_lose6,right0_lose6,right0_lose6,left0_lose6,left0_lose6,
right100_win7,right100_win7, left100_win7,left100_win7, right50_win7, left50_win7, right50_lose7, left50_lose7,right0_lose7,right0_lose7,left0_lose7,left0_lose7,
right100_win8,right100_win8, left100_win8,left100_win8, right50_win8, left50_win8, right50_lose8, left50_lose8,right0_lose8,right0_lose8,left0_lose8,left0_lose8,
right100_win9,right100_win9, left100_win9,left100_win9, right50_win9, left50_win9, right50_lose9, left50_lose9,right0_lose9,right0_lose9,left0_lose9,left0_lose9,
right100_win10,right100_win10, left100_win10,left100_win10, right50_win10, left50_win10, right50_lose10, left50_lose10,right0_lose10,right0_lose10,left0_lose10,left0_lose10)
; 61 = rest1]


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
That's not going to be easy if at all possible. My first intuition is that you'll have to use a set of <list>s to do the trial selection. Encode the trialtypes in one main <list> numerically. Place constraints on that list's selection as needed (e.g. using /maxrunsize).

Have another <list> linked to the above that contains the actual trial objects. Then do:

<block myblock>
/ trials = [1-9=list.triallist]
</block>

<list typelist>
/ items = (1,1,1,2,2,2,3,3,3)
/ maxrunsize = 1
</list>

<list triallist>
/ items = (trial.x_a,trial.y_a,trial.z_a,trial.x_b,trial.y_b,trial.z_b,trial.x_c,trial.y_c,trial.z_c)
/ selectionmode = list.typelist.nextindex
</list>

<trial x_a>
/ trialduration = 50
/ validresponse = (0)
</trial>

<trial x_b>
/ trialduration = 50
/ validresponse = (0)
</trial>

<trial x_c>
/ trialduration = 50
/ validresponse = (0)
</trial>

<trial y_a>
/ trialduration = 50
/ validresponse = (0)
</trial>

<trial y_b>
/ trialduration = 50
/ validresponse = (0)
</trial>

<trial y_c>
/ trialduration = 50
/ validresponse = (0)
</trial>

<trial z_a>
/ trialduration = 50
/ validresponse = (0)
</trial>

<trial z_b>
/ trialduration = 50
/ validresponse = (0)
</trial>

<trial z_c>
/ trialduration = 50
/ validresponse = (0)
</trial>

GO

Merge Selected

Merge into selected topic...



Merge into merge target...



Merge into a specific topic ID...




Reading This Topic

Explore
Messages
Mentions
Search