Experiment doesn't work

débutanteopensesame

Hi,

I have to do an experiment for my last year of study and I’m lost... I also have a problem with my experience, I can’t create a sound variable with different sound files and python script crashes every time. I can't reach her because I am receiving error code 413. If anyone wants to look at it I can e-mail it, that would be super nice!

lvanderlinden

Hi @débutanteopensesame ,

Could you upload your experiment here by clicking the paperclip icon?

And will you be running your experiment online or locally?

Cheers,

Lotje

débutanteopensesame

I'dI can't download it here because the following message is displayed:

request failed with status code 413. 

How to transfer it to you otherwise?

I'd like to run it online.

lvanderlinden

Hi @débutanteopensesame , does it help if you empty your file pool? Otherwise you can send it to me by email: l.vanderlinden @ cogsci.nl

débutanteopensesame

I sent you an e-mail :)

lvanderlinden

Hi @débutanteopensesame ,

Thanks for the experimental script. How should the sounds be matched to the words? Does each sound go together with a specific word? Or should the sounds be chosen randomly?

And should the sound be played during the word that is presented, or before (i.e. during the preceding fixation point)?

And what should be happening in the part that is called "planexperimental"? Coroutines are not supported in OSWeb (and neither are form_text_input items, unfortunately)...

Cheers,

Lotje

débutanteopensesame

Hi,

The experience corresponds to the three training phases at the start: point of fixation (time 0 seconds), followed by a sound and words. The words are listed (12 words) and there is a reminder at the end. The sounds and the lists are presented in block: three blocks made up of 10 lists of words each, which makes a first block with a sound (for example iso). After 10 word lists a second block with another type of sound and a third block with another sound. The iso and non-iso sounds would, in the best case, be presented randomly for each list (example: list 1 -> its iso 1, list 2-> its iso 2 etc).

Hoping to be clear :)

débutanteopensesame

In the "planexperimental" phase, there are presentation of words lists (30 in total) divided into 3 blocks with the three different types of sounds.

Ok, so I should do my experiment locally or we would have to find a way to do without

lvanderlinden

Re-bonjour @débutanteopensesame :)

The answers below consider the practice phase of your experiment:

I can’t create a sound variable with different sound files

You could define the sound variables one level higher than the level where you defined the words, as in the example below.

The iso and non-iso sounds would, in the best case, be presented randomly for each list

To achieve this:

• Set the order of the loop where you define the sounds to "random"
• Set the repeat value to 0,15. With 20 rows in the loo, this makes sure that only 3 of them will be

Next, you can use the square-bracket notation in the sampler item to play the sound belonging to the current list like so:

Finally, to make sure that only one list of words, and only one interference task are run in a given block, you could use Run-if statements like so:

(Note that Python starts counting at 0, which is why the first time the block_sequence is run, the corresponding counter is 0.)

I uploaded an example file (but with an empty file pool).

6989.osexp

And python script crashes every time

I noticed... Would it help if you converted your .wav files to .ogg? See also here:

• https://forum.cogsci.nl/discussion/4722/python-crashing-when-running-experiment

Finally, as I pointed out earlier, form_text_input items are not supported in OSWeb.

• https://osdoc.cogsci.nl/3.3/manual/osweb/osweb/#supported-functionality

You can still collect text input, but it is not as intuitive as it is when running OpenSesame locally (and using form items):

• https://forum.cogsci.nl/discussion/5876/collecting-multiple-character-input-in-osweb-e-g-student-numbers#latest

Hope this helps! Let us know if you also need help creating the experimental part.

Cheers,

Lotje