[solved] Go backwards / forwards within a sequence and record all of the keyboard response times

Peter

Hi, Sebastiaan,

Just similar with David’s question in http://forum.cogsci.nl/index.php?p=/discussion/119/open-backwards-within-a-sequence. In my experiment, I want to present 50 pictures in a sequential order. The participants can go backwards / forwards within the sequence by pressing a specific key. Besides, I also want to know all of the response times, so I need to see all records in the logger.

Although l had studied the example that you made in http://forum.cogsci.nl/index.php?p=/discussion/comment/215, but I still have no idea how to use it. Could you teach me how to do this with inline code, please? Thank you so much.

Best regards,

Peter

sebastiaan

Hi Peter,

This is not trivial, as you have experienced, but I thought of a better and simpler way to do it.

First, don't put trial_sequence in a loop, but at the start of the experiment with 'run_if = never' or in the 'unused items' bin (or anywhere, as long as it's not automatically called). Then insert the script below instead of a block loop. This will call the trial_sequence and advance or go back based on the response of the user. You will also manually have to set the variables, because there's no loop table to help you. For more info, see the comments in the script.

# Uncomment for 0.25 and before
# exp = self.experiment

# Create a list of conditions, such as picture names. Change
# based on your experiment.
condList = ['pic1', 'pic2', 'pic3', 'pic4', 'pic5'] 

# Loop until the last trial has been done
trialId = 0
while trialId < len(condList):

    # Explicitly set the appropriate condition based on the
    # condition list and the trial id. Change the variable
    # name from 'pic' based on your experiment.
    exp.set('pic', condList[trialId])
    
    # Prepare and run the trial_sequence. Change 'trial_sequence'
    # based on the name of the item that you want to run.
    exp.items['trial_sequence'].prepare()
    exp.items['trial_sequence'].run()
    
    # If 'response' was left, go back one trials, otherwise
    # advance one trial. Change this criterion based on your
    # experiment.
    if self.get('response') == 'left':
        trialId -= 1
    else:
        trialId += 1
        
    # Negative trial IDs are not valid
    if trialId < 0:
        trialId = 0

Hope this gets you started!

Cheers,
Sebastiaan

Peter

Thank you Sebastiaan!

You’re so nice and replied quickly. However, I tried many times, it still didn’t work. The debug window showed lots of errors:

Error: Inline script error
In: trial_sequence (run phase)
File "libopensesame\inline_script.pyo", line 123, in run

Python traceback:
inline_error: Error: Inline script error
In: trial_sequence (run phase)
File "libopensesame\inline_script.pyo", line 123, in run

I didn’t know What I have done wrong. Please help me. By the way I'm using OpenSesame 0.26.

Thanks again.

Best regards,

Peter

sebastiaan

I think something must have gone wrong with copy-pasting the code into OpenSesame's inline_script editor. Please make sure that all the indentation etc. is correct. If it doesn't work you can just copy the relevant bits by hand. The script itself should be fine.

Cheers!

Peter

[ITA] Someone expressed an intent to answer your post. This means that person intents to answer your question (very) soon, but doesn't have the time right now.

Peter

Hi Sebastiaan,

Thank you so much. When I copied the relevant bits by hand, it seemed to work. However, there is another problem with variable. I don’t know how to set it without loop table. Could you please take a look at the simple example below? Tell me what I have done wrong. Thank you.

Direct link (save as .opensesame file) :
https://docs.google.com/open?id=0B30WJRj7lOisTHRFdk04blQwODA

Best regards,

Peter

sebastiaan

Hi Peter,

Right, I can see why this would confuse you (it confused me too for a bit). The thing is that OpenSesame tries to prepare the trial_sequence, even though it will not be executed ('run if = never'). Because the preparation occurs before the inline_script is executed (which has only code for the 'run' phase, which comes after the 'prepare' phase) the variable 'pic' has not yet been set when OpenSesame tries to prepare the sketchpad. Hence you get an error. Does this make sense?

There are (at least) two ways to solve this:

• Give 'pic' a default value at the start of the experiment, for example by opening the general script (general tab → show script editor) and adding something like this to the top of the script (and press apply):

set pic 1

• Delete 'trial_sequence' so that it's in the Unused items bin. Unused items are never prepared. But don't empty the Unused items!

Good luck.

Cheers,
Sebastiaan

Peter

Hi Sebastiaan,

It works! It works! I really appreciate your help! The first way is great to my experiment. Thank you!

Best regards,

Peter