|
|
Post by estocastico on Aug 21, 2020 1:38:54 GMT
Hello there! I hope you all are doing well. I'm in the final stage of building an experiment, and there is one thing a cannot find a way to set it up. I have a table with 8 unknown objects that appear in 4 different contexts each (column 1 below), 32 images in total, that are associated with a pseudoword (column 2). I want the participants to see each image and listen to the associated audio, but I don't want them to hear the same audio consecutively.I know that I can manually randomize my list (table) and use the "fixed" command in the block, but this may be a problem for me, because my experiment involves word learning, and there is the possibility that the participants learn the word that appear first better than the last, and I want to avoid that. My code look like this:
table learning
object1context1 pseudoword1
object1context2 pseudoword1
object1context3 pseudoword1
object1context4 pseudoword1
object2context1 pseudoword2
object2context2 pseudoword2
object2context3 pseudoword2
object2context4 pseudoword2
object3context1 pseudoword3
object3context2 pseudoword3
object3context3 pseudoword3
object3context4 pseudoword3
...
object8context1 pseudoword8
object8context2 pseudoword8
object8context3 pseudoword8
object8context4 pseudoword8
task learning
table learning
show bitmap fixpoint
delay 500
show bitmap @1
delay 1000
sound look
delay 1500
sound @2
delay 1500
clear -1
block learning
message instructions
tasklist
learning 32 all_before_repeat
end
It works really well, but I don't know the way to set it up without the possibility of playing the same sound (showing the same object) twice in a row. Is there an instruction/command that I'm missing? Can I do that here? How do I configure it? Thanks in advance. Best, |
|
|
|
Post by alexandra (AEC) on Aug 21, 2020 9:24:07 GMT
Hi there,
off the top of my head I would say that you could try using global variables and local variables. For example, at the start of your task you set a local variable to the sound(number) of the line that is called:
set $mysound @2
and then you compare it to &priorsound using if
if $mysound != &priorsound [... your task code] set &priorsound $mysound fi
At the end of your task, if prior sound and current sound were different, you want to set prior sound now to the sound that was played. And this goes on....
You probably want to know what sounds were played I guess? If so, than you will need to save the variables at the and of your task. For example:
save $mysound
It could be useful to save both &priorsound and $mysound during your design stage just to check that this actually works and you do not get the same sound directly after each other.
|
|
|
|
Post by estocastico on Aug 21, 2020 18:34:05 GMT
Hi Alexandra! Thank you for your help. I tried and tested what you said. My task code ended looking like this:
task learning
table learning
set $mysound @2
if $mysound != &priorsound
show bitmap fixpoint
delay 500
show bitmap @1
delay 1000
sound mira
delay 1000
sound @2
delay 2000
clear -1
set &priorsound $mysound
fi
save $mysound
It technically works. When I ran the experiment, there were no repetitions of the same sound consecutively, but the data showed that there were some repetitions. I ran the experiment again and manually wrote down the sounds, and again there were no repetition, but there were less trials than I've programed. The data showed repetitions and all the trials I want.
When comparing the data and what I wrote, the number of consecutive repetitions is the same as the trials that are missing. For example, when the data shows 3 consecutive repetition of the pseudoword number 7, I ended having 2 trials less than expected.
It seems that when a sound is the same as the previous one, it just get skipped. It doesn't get shown, but still count as a trial.
Thank you for your help again!
|
|
|
|
Post by alexandra (AEC) on Aug 24, 2020 16:01:28 GMT
Yes, you observed this correctly. For one trial one row of your table is read for all information provided in that row. An if statement like it was used here just means that the trial is "happening" but nothing is shown, when the prior sound was the same as the one now provided in @2. If you need a certain amount of trials each time or if each row of your table needs to appear at least once than it gets a bit more complicated and you will need to use a lot more variables and/or create additional tables and/or run another block.
|
|
|
|
Post by PsyToolkit on Aug 24, 2020 23:09:15 GMT
In the upcoming version, such issues will be easier, because you can use array variables. Depending on how long you can wait, you can see if that can work for you (you can have a peek at it via www.psytoolkit.org/c/3.2.0/
It is still being tested, but most of it seems to work just fine.
|
|
|
|
Post by estocastico on Aug 27, 2020 20:12:34 GMT
Yes, you observed this correctly. For one trial one row of your table is read for all information provided in that row. An if statement like it was used here just means that the trial is "happening" but nothing is shown, when the prior sound was the same as the one now provided in @2. If you need a certain amount of trials each time or if each row of your table needs to appear at least once than it gets a bit more complicated and you will need to use a lot more variables and/or create additional tables and/or run another block.
I'm just getting the hang of the set function. Just recently undertstood what the instruction says and is clear to me now. And yes, I need that each row of the table appears once. I've thinking on how to program it, but I'm a little stuck for now.
On the other hand, where can I read more about local and global variables? Where can I find them? Didn't know somethig like "&priorsound" existed, I've been searching about or others similar variables, but could't find them.
In the upcoming version, such issues will be easier, because you can use array variables. Depending on how long you can wait, you can see if that can work for you (you can have a peek at it via www.psytoolkit.org/c/3.2.0/It is still being tested, but most of it seems to work just fine.
Wonderful! I'll give them a try (if I get to master those functions haha. I will).
Thank you, both!
|
|
|
|
Post by alexandra (AEC) on Aug 28, 2020 8:58:38 GMT
You can find more information on variables and other useful information on PsyToolkit syntax for experiments here.
& is used for a global variable while $ is used for a local variable. I just named it "priorsound" but you could name it any other way.
|
|
