maxrunsize in list, streaks at the end of a block


Author
Message
nrouhani
nrouhani
Distinguished Member (3.5K reputation)Distinguished Member (3.5K reputation)Distinguished Member (3.5K reputation)Distinguished Member (3.5K reputation)Distinguished Member (3.5K reputation)Distinguished Member (3.5K reputation)Distinguished Member (3.5K reputation)Distinguished Member (3.5K reputation)Distinguished Member (3.5K reputation)
Group: Forum Members
Posts: 83, Visits: 261
hello, 
in my experiment, two trial types are sampled throughout a block, 16 times each. I do not want more than two of the same trial type to be sampled in a sequence. the problem I am running into is that when the maximum number of one trial type has been reached (16), there is a long streak of the other trial type until the end of the block. Is there any way to help or prevent this? 

Thanks! 

<list triallist>
/ items = (trial.cue_indoor, trial.cue_outdoor)
/ poolsize = 32
/ maxrunsize = 2
</list>


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: 105K
nrouhani - Friday, September 15, 2017
hello, 
in my experiment, two trial types are sampled throughout a block, 16 times each. I do not want more than two of the same trial type to be sampled in a sequence. the problem I am running into is that when the maximum number of one trial type has been reached (16), there is a long streak of the other trial type until the end of the block. Is there any way to help or prevent this? 

Thanks! 

<list triallist>
/ items = (trial.cue_indoor, trial.cue_outdoor)
/ poolsize = 32
/ maxrunsize = 2
</list>


The only way to prevent this would be to define one (or several) fixed sequences that satisfy the given constraint(s) (here: no more than 2 repetitions). The sampling from list elements is conducted on-the-fly: Taking your <list> as example, you can imagine it as drawing from a deck of 32 cards. 16 of those cards are labeled "indoor", the other 16 are labeled "outdoor". You're drawing cards from the list randomly (without replacement). Each trial can be understood as drawing one random card from the deck. So, suppose for the 1st trial you draw a card labeled "outdoor". The 2nd card drawn is also labeled "outdoor." Now, due to /maxrunsize=2, if the 3rd card drawn is _also_ "outdoor", Inquisit puts that card back in the deck, and selects a different card. This continues until it draws a card that does _not_ violate the constraint, i.e. one labeled "indoor." The 4th card, however, can again be an "outdoor" card. It's easy to see that one might run into situations where as fewer and fewer cards are left in the deck, the only cards remaining cannot satisfy the constraint anymore.

See https://www.millisecond.com/forums/Topic6419.aspx for some additional thoughts, the scenario outlined above falls under section (II).

GO

Merge Selected

Merge into selected topic...



Merge into merge target...



Merge into a specific topic ID...




Reading This Topic

Explore
Messages
Mentions
Search