Moving back and forth between routines

Builder
21

The problem appeared when I adapted your example to my experiment. I added a end page that is not part of the loop, but when I get to the last question, it goes back to the first image. It is like the loop is not ending. I can go back to the pics when I’m in the questions but I can’t move to the end page.

22

Hi Jan,
Your demo helped me a lot while doing my experiment and I was able to use it for both my image and text condition. However, right now, the order is Image -> Text. Is there any way I can randomize this order and make it Text -> Image for some participants. I saw similar topics here saying that we should try to minimize routines and put both conditions in one routine, however I also have questions and the repetition of pictures in my experiment so I’m not sure how should I proceed.
Thank you in advance for any tips you might share.

23

Sorry, could you explain this in a bit more detail? Would the idea be that some participants simply skip the first set of pictures (so, the order is questions → pictures → questions etc.), but everything else basically stays the same?
Or would this be a complete reversal? That is, participants can go back to questions while they look at the pictures and if they don’t decide to go back to the questions, the loop ends after they have seen all the pictures?

24

Sorry, I should have explained better.
The images are actually only one of the conditions for this experiment. The other condition are sentences that are presented just like the pictures, in sequence and have specific questions that the participants need to answer. I used the demo your sent me for both the pictures and the sentences. What I want to do now is to counterbalance the order of the two conditions so that for some participants, they see pictures(answer the questions) -> sentences(answer the questions) and other participants see sentences(answer the questions) -> pictures(answer the questions).
I searched for some ways to do that and they all talked about blocks but I’m not sure if it would be possible because of the different questions that are attached to each condition and the repeated conditions(participants going back to check the conditions when they are answering the questions -your demo).

25

Ah, I see. So, how do you currently switch from pictures to sentences? (Perhaps add a screenshot of your flow. Or upload a simplified version of your experiment.) Also, how are your input files set up?

26

Hi Jan,
I’m uploading a simplified version so you can see what I mean.
Basically, I created another loop with routines just for the text, exactly like the demo you sent me for the images.
Like I said, what I’m trying to do is to counterbalance the order image/text are presented to participants. I read from the forum many people saying I should keep everything in one block, but I have different instructions and questions for image and text. Furthermore, in the text condition I’m trying to add another variable(Text2). That is, the sentences that would be used in the text conditions would vary according to participants. Participant 1 would get text and participant 2 would get Text2, for instance. I created a topic for this part here, but I’m still trying to fix it.
I know it is a lot, but I would appreciate any insight on this because I’m really lost on how to do it.
repeatexp.psyexp (32.1 KB) images.xlsx (8.5 KB) questions.xlsx (8.5 KB) questions2.xlsx (8.5 KB) text.xlsx (8.4 KB) Text2.xlsx (8.4 KB)

27

Hi,

Please find a minimal solution attached. It uses an outer loop for counterbalancing. The startup dialogue asks which order you would like to run. If you enter ti, the order will be text→image; it will give you image→text. The counterbalancing loop has two repetitions. Say, you go for it, then:

  • First iteration of the counterbalancing loop (i.e., counterbalLoop.thisN == 0):
    • nReps for textLoop will be 0 (i.e., element 0 in list textRep)
    • nReps for imageLoop will be 1 (i.e., element 0 in list imRep)
  • Second iteration of the counterbalancing loop (i.e., counterbalLoop.thisN == 1):
    • nReps for textLoop will be 1 (i.e., element 1 in list textRep)
    • nReps for iamgeLoop will be 0 (i.e., element 1 in list imRep)

Jan

counterbalance.psyexp (10.3 KB)

1 Like
28

That is amazing Jan.
Thank you so much, you’re always really helpful!
Wishing you the best!