Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Supported by

Constraining Task Switches: Switch in 75% of Trials, Constrain Rare Non-Switch-Condition to Maxrep

edited October 2019 in OpenSesame

Beautiful community!

For days I've been trying to get my head around this, and have found numerous posts which looked promising, but after further reflection didn't really answer my problem, which I again turn to you with a new thread, sorry about that.

I am trying to reproduce this experiment from Fröber & Dreisbach (2017, 10.1016/j.cognition.2017.01.024), namely the Task Switching part. The task goes like this: above or below a fixation cross, either a number or a letter can appear (it's counterbalanced, to keep it simple, let's just say the number is above, the letter below). The subject's task is to decide, whether the number is greater or smaller than 153, or whether the letter is closer to A or to Z, respectively. The number stimuli are 125, 132, 139, 146 (correct answer: y), 160, 167, 174, 181 (correct answer: x). The letters B, D, F, H (correct answer: n), S, U, W, Y (correct answer, m).

The problem is the switching part (again, here's several versions, but that is not the problem, so I'll stick to one example): let's say the switch is supposed to occur in 75% of trials, and only in 25% of the trials there is the same judgement (letters resp. numbers) to be made in two consecutive trials. Furthermore, the rare condition (here: repetition of judgment) cannot occur more than once in a row. Also, no direct stimulus repetitions are allowed.

I list here a number of resources I have found so far that seem to point in the right direction, but I'm really struggling to adapt those to my case, especially since some reference attached experiment files that sadly aren't there anymore:

As always, any help is greatly appreciated.

Best, Vincent


  • Hi Vincent,

    this may help you out. Of course, it requires inline scripting.

    Whether it's considered a sequence or a (task) switch does not matter. What counts is that we all want to constrain the alternation from one trial to the other, but keep it somehow random. ?

    With the script you get a different order for each participant, but with the same constraints. You may adopt the constraints to your needs, of course.



Sign In or Register to comment.