OSWeb Loop/Block Convertion not working properly.

Comments

• VictorVarela

  January 2021

  

  Main Loop contains a list of 5 blocks (N, W, T, A)

  And Block_loop contains 100 images
• lvanderlinden

  January 2021

  Hi @VictorVarela

  
  

  Break-if statements should work in (the latest versions of) OSWeb. Could you share your experiment by another medium, such as pastebin, so that we can have a closer look?

  To paste the experimental code, you can click the very first item in the overview area (with the OpenSesame item), click General script and copy the code.

  
  

  (Or you could try the upload function a couple of more times, perhaps it will work now.)

  
  

  Cheers,

  
  

  Lotje

  Did you like my answer? Feel free to
• VictorVarela

  January 2021

  Okay I guess this is what you asked for

  
  

  https://pastebin.com/t7ZZAADr

  
  

  Thanks @lvanderlinden, as always :)
• lvanderlinden

  January 2021

  Hi @VictorVarela ,

  
  

  Unfortunately, the "resume-after" option is not supported in OSWeb yet. (So there's no problem with your break-if statement itself.)

  
  

  Is it absolutely necessary for your experiment to have the breaks? And for the trials to be completely random? (If not, you could create 5 blocks of 20 trials, of course).

  If the breaks and the random pulling are indeed necessary, I guess we should implement it in JavaScript. I can look into it, but I'm not particularly good with JS myself... ;)

  
  

  Cheers,

  
  

  Lotje

  Did you like my answer? Feel free to
• VictorVarela

  January 2021

  Dammit!!

  Well, at least now we know where the problem is... I will make it sequential if it is not yet implemented :(

  
  

  Thanks a lot @lvanderlinden
• mgade

  February 2023

  Hi, I have a related issue. I have an experiment with two blocks in the offline version which runs smoothly but as soon as I turn to the online version (i.e. even when just testing online version, the second block is always skipped). I have this issue with two completely different experiments using OpenSesame in its latest version and Firefox or Chrome as default browser, one experiment is attached as example, but the issue replicates in another study. My OS is Windows 10.

  I would be very thankful for any suggestions, best Miriam

  BAHerkensV7.osexp

  
  
• lvanderlinden

  February 2023

  Hi @mgade ,

  
  

  I noticed that you have some sequence items within other sequence items. You should, however, always wrap a loop item around a sequence item (except for the one called 'experiment' at the very top of the overview area. Could you try whether this solves the issue?

  
  

  Cheers,

  
  

  Lotje

  Did you like my answer? Feel free to
• mgade

  February 2023

  Hi,

  thanks for your comment.

  So you mean never call a sequence which not embedded in a loop? I was figuring out something in the same direction and wrapped a sequence around my two blocks which actually made things worse, obviously getting the hierarchy of sequences and loops wrong. I will play around tomorrow and will let you know - regards Miriam
• lvanderlinden

  February 2023

  HI @mgade ,

  
  

  O, I see what you mean! The structure seems overly complex! ;)

  
  

  What is it that you want to achieve with all those nested pairs of loop and sequence items? Is it just to counter balance the order of the written-word and the tone blocks between participants?

  
  

  Cheers,

  Lotje

  Did you like my answer? Feel free to
• mgade

  February 2023

  hi Lotje,

  exactly that's what I intended and I have no idea how to solve this within a loop. Also my experience with online experiments suggests that shorter blocks and more breaks are better to keep people going :-)
• lvanderlinden

  February 2023 edited February 2023

  Hi @mgade ,

  
  

  I tried to clean up the experiment by using *linked* copies of as many items as possible, because having *unlinked* copies or doing the same work multiple times is not very good practice (and error prone). Also, I put the questions of the initial questionnaires inside a loop`-`sequence structure, with as a main benefit that you do not have to use multiple logger items (which is also not good practice).

  
  

  In understand what you intended to do by counterbalancing the order of the two types of blocks on the basis of subject number. This implementation, however, only works offline, where you, as an experimenter, indicate the subject number. In contrast, in OSWeb/JATOS, the subject number is randomly chosen (by default the number 0 or 1). It is therefore *not* guaranteed that you end up with an equal number of participants in each group.

  
  

  Do you see the difference? You therefore have two options:

  
  

  Option 1

  Set the order of the called "experimental_loop" to random. In this case there is guarantee of an exact match in number of participants (for example, you end up with 46 participants who first did the pitch version and then the word version vs 54 participants who did the blocks in the opposite order).

  

  Option 2

  Set the order of the aforementioned to sequential and make two identical versions of the task except that you manually change the order of the block type. (Do this only when you are completely happy with your experiment and only have to make this final change.)

  

  

  
  

  Hope this helps!

  
  

  Cheers,

  
  

  Lotje

  6816.osexp

  
  

  Did you like my answer? Feel free to
• mgade

  February 2023

  Hi Lotje,

  thanks I will use your draft to create my versions and thanks for cleaning up the code, I found it awfully messy myself. regards