Howdy, Stranger!

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

Supported by

WSCT Tutorial

Hi all

I am trying to use OSweb for the first time, and I implemented a short version of the WSCT , using only the shape-rule (https://osdoc.cogsci.nl/3.3/tutorials/wcst/) and 6 trials.

Here is the .osexp file: https://drive.google.com/file/d/1TyJgWrj-lcaxSCTtXLnR-6COt9xP04TD/view?usp=sharing


The task works fine when I test it in OpenSesame, and results csv file is ok.

Then I uploaded the task in Jatos, and run it online. It seems to work fine: https://experiments.jatos.org/publix/1535/start?batchId=1831&generalMultiple


But, when I check my results (after converting them back to xlsx format), "None" is registered as a response, and the response time recorded is not the real one (I delayed my keypress to see if it would be recorded, but rts are always around 510 ms). Keep in mind that the "correct" variable is registered as it should!!!.

Here is a snapshot of the xlsx file:


Can someone please suggest what the problem might be?


Thank you

Fotis

Comments

  • edited November 2020

    Hi @fotisfotiadis ,


    Not sure why this discrepancy between OpenSesame and OSWeb occurs. In any case, you can solve the issue as follows:


    • Use only one keyboard_response item, call it "keyboard_response" (which is its name by default), do not set the correct response here (but see below), and set its Run-if statement to "always"
    • Use the variable "response_keyboard_response" to inspect the responses (and add it to your logger)
    • In order for OpenSesame to determine whether a response is correct or not, you can add a column called "correct_response" (case sensitive) to the block_loop and define the correct response per trial over here

    I attached a modified script. (I hope I defined the "correct_response" variable correctly...)


    Hope this helps!


    Cheers,


    Lotje

    Did you like my answer? Feel free to Buy Me A Coffee :)

  • On a different note: Shouldn't the order of the trials in the block loop be set to "sequential" instead of "random"? Since the point of the WCST is for participants to guess (new) rules on the basis of a sequence of trials?

    Did you like my answer? Feel free to Buy Me A Coffee :)

  • Dear Lotje

    Thank you so much for your time. I wouldn't expect anyone to modify the script for me. Again, thank you.

    Just for the record, you are correct about the sequence of trials, but the correct variable is a bit more complicated. Luckily, the java script is here: https://osf.io/f5er2/


    One more question if I may, please.

    I discovered that I am unable to display Greek characters in a feedback item . No problem with a sketchpad though (I am only talking about the OSWB feature, there is no problem with the desktop version).

    Eg., in this stroop task (https://drive.google.com/file/d/1lJx5ad_Fg6FT9DrKnNv6Pt6r6frp0axY/view?usp=sharing), feedback is the words "correct" and "wrong" in Greek, which cause the Uncaught TypeError: Cannot read property 'replace' of undefined error. (If you wish to try it out, press 1 for red color, 2 for yellow, 3 for green)

    Is this a known issue?


    Again, thank you for your time,

    Fotis

  • edited November 2020

    Hi @fotisfotiadis ,


    I see what you mean. I don't know what's causing this issue. I'll check with @sebastiaan to see whether this should be filed as an issue on OSWeb's github page:



    Cheers,


    Lotje

    Did you like my answer? Feel free to Buy Me A Coffee :)

  • Hi @fotisfotiadis and @lvanderlinden ,

    The issue appears to be that OSWeb doesn't correctly parse non-Latin strings if they're not quoted (and single-word text elements aren't quoted by OpenSesame). I filed an issue for that here. For now, a simple workaround would be to add a single space before and after the words on the feedback item. If you look at the script, you'll see that this causes the text to be quoted, thus resolving the issue.

    — Sebastiaan

Sign In or Register to comment.