[wishlist] What would you like to see in OpenSesame?

labovich

Hi
Just started using the software. Really loving it!! you are doing a great work!

• It would be nice to have option for nested tables (without writing additional code)
• It will be useful to be able to change the order of the columns in a loop table, and to delete more than one column at a time...

Thanks again!

refuzee

ctrl+z

export with all dependencies (vlc player, python...) as .exe file.

standardized way to handle questions and texts.

random durations withouth inline code.

best, daniel

bs0dcv

This may be more of an add-in than a change to OpenSesame but a way to link the LabChart software from AD Instruments that measures physiological responses would be really good. Synchronising timing and combining data in the logger would open up more experimental testing possibilities.

mariosav

I have taught OpenSesame as part of an undergraduate class on Experimental Psychology which solved a lot of the headaches I used to have when using E-Prime in the past (e.g., handing out license keys to students, letting them take they keys home, windows-only etc). And students loved it! The only thing that could make my life easier (i.e., avoid coding in an undergrad class) is "nested lists" functionality. And a late thing in my wish-list: drop-down menus to append new/existing items! Such a convenient feature...why let it go??

Jarik

A "StartHere" item would be very nice when building a big experiment consisting of many parts. One could drag it anywhere in the overview area to make the experiment start at that specific point to test only that part of the experiment. To prevent accidental skipping of the beginning of an experiment start here might only be activated when using the 'Quick run' option.

Best, Jarik

refuzee

question randomization in forms as stated here:

http://forum.cogsci.nl/index.php?p=/discussion/comment/5118#Comment_5118

elaci0

I see touch_response works with cells of screen decided upon row/column.

Given that it's quite difficult to guess if my "Image" is in that cell, could be possible to add a sort of superimposed grid which shows how the screen would be divided using X rows and Y columns?

Xophmeister

It would be nice to add some reflection/introspection/metaprogramming capabilities, via Python inline scripts, such that you could manipulate items in the experiment sequence at runtime.

For example, I'm rolling my own randomisation scheme and then using a simple loop with a hardcoded sequence that will later refer to my randomised list. That hardcoded sequence can't be changed dynamically to reflect the bounds of my list (which can be manipulated with the full power of Python; e.g., fetching data from a DB).

Thanks;
C

Michif

Hi,

not sure whether that has been wished already but it would be very handy if the text item would have a "go back" option. Kind of restarting the current e.g. sequence item.

Oftentimes, participants just continue to quickly to read through the content of the text items containing experimental instructions. It happens now and then that they skip on text item too quickly or by accident. Usually we then have to press ESC and restart the script. That is only if we specifically ask for whether something happened because sometimes participants would just continue and do not tell us that they skipped one part of the instructions.

Cheers,
M

ArathyPuthillam

Hello.
Is there a way to add responses (of the participant) through speech?

Knante

Hi,

I'm not sure if these are specifically mentioned in the posts above but i would like to see:

• Unlink copy option (when an existing item has been copied)
• Improved grid in the touch-response item, for example being able to set the coordinates of the allowed response fields and the fields that should be blocked
• Similar to Elacio a few posts above: a superimposed grid to check whether presented stimuli are overlapping with the response grid

Thanks,

Laurent

eugenia

Hallo,

I am new to opensesame and I am not sure it is the right place where to post.

By the way i would like to ask you whether it is possible by using open-sesame on the tablet to drag and drop figures on the screen by using fingers,
it would be useful for me because I am programming an experiment for kids.

thanks a lot,

Eugenia.

klsou

Hello,

I am would like to use opensesame to do some experiment which involves displaying video. However, I realized that there is no way to change the width and height of the video in the programme. So I would like to add this function into this wish-bucket.

Kalon

refuzee

We would very much love a do answer check option, to control if users really did select a choice or entered a text. This somehow was questionned a couple of times in the forum already and is only solved by scripting and adapting for each experiment

zennie

Simple method for collecting RT for key releases, instead of key presses.

amihai_gottlieb

Would it be possible to allows to collapse items in the overview and leave them collapsed until you want to expand them again? As it is, I can collapse all those loops, but the application auto-expands them on the next command of any kind.

mfp49

Many thanks.
I would be very glad if the OpenSesame could support presentation of the Farsi fonts and presentation.

DCube

Hi,

I mentioned this in another thread. It would be nice if Open Sesame's first item would be a loop rather than the experiment. That way the experiment would be always on unless otherwise told to (i.e., by a special key combination) and participants would not get to see the insides of the Open Sesame experiment.

This could remove the curious participants' search that may send them to areas in the experiment with sensitive data either in scripts (that they may mingle and change it just for fun) or to areas where the answers of the tests may be.

DCube

Jarik

Hi @DCube,

Guess that is what OpenSesame Run is for:

• http://osdoc.cogsci.nl/3.0/usage/opensesamerun/

Note the link is to the 3.0 documentation not the latest 3.1 documentation but I think OpenSesame Run still comes with OpenSesame like it did in version 3.0.

@sebastiaan am I not looking hard enough or is there no OpenSesame Run in the 3.1 documentation?

Best,
Jarik

sebastiaan

@sebastiaan am I not looking hard enough or is there no OpenSesame Run in the 3.1 documentation?

Oops. It got lost in the re-organization. I'll re-add it soon.

Daniel

@amihai_gottlieb that has been one of my gripes for a while now too. I plan on taking a look at this in the near future. It shouldn't be that difficult as we've already solved a similar problem in the OSF explorer.

TomArmstrong

There is already a great thread on work-arounds, but I second the request for nested lists (in line with E-Prime), so that undergrads who can't code can do more with the software. Also, something like the clock.scale function in E-Prime, which allows you to speed up the experiment for testing, would be awesome. Finally, expanding PyGaze to include the Tobii Eye X (affordable) eye tracker, given the limited volume of Eye Tribe eye trackers, would be great. Thanks for considering the suggestions!

sebastiaan

@TomArmstrong

There is already a great thread on work-arounds, but I second the request for nested lists (in line with E-Prime), so that undergrads who can't code can do more with the software.

As of OpenSesame 3.1, there are several advanced-loop operations. One of those is the horizontal shuffle, shuffle_horiz), which does exactly what most people (to my understanding) use nested lists for in E-Prime.

• http://osdoc.cogsci.nl/3.1/manual/structure/loop/#shuffle_horiz

As it happens, I just posted a video tutorial on exactly this:

Does that cover nested lists, or is there still cases that you cannot implement with shuffle_horiz or any of the other advanced-loop operations?

Also, something like the clock.scale function in E-Prime, which allows you to speed up the experiment for testing, would be awesome.

That's an interesting suggestion. If you add the following to the prepare phase of an inline_script at the very start of the experiment, most things will speed up:

import functools

SPEEDUP = 10 # Speed up things by a factor of 10

def scaled_sleep(ms, sleep_fnc=None, speedup=1):

    sleep_fnc(int(ms / speedup))

def scaled_time(time_fnc, speedup=1):

    return time_fnc() * speedup

clock.sleep = functools.partial(scaled_sleep, sleep_fnc=clock.sleep,
    speedup=SPEEDUP)
clock.time = functools.partial(scaled_time, time_fnc=clock.time,
    speedup=SPEEDUP)

However, this doesn't speed up everything. For example, keyboard timeouts will stay as they were.

Cheers,
Sebastiaan

DahmSF

Dear OpenSesame Crew,

I would like to save my experiment in a way that you run it directly by double clicking on the file.
I will install it on the participants laptops. And i am worried that they might change something in the setting/programming. Or select the wrong start button.
I know that there is another way to start the experiment. However, this is quite complicated. I can do that, but for the participants it would be easier to open a file with a double click.
In Eprime they integrated such a tool, because of licencing. However, as explained it would be useful in OS as well.

I appreciate all you have already implemented. Great job!
Stephan

sebastiaan

Hi Stephan,

Sounds like you need opensesamerun:

• http://osdoc.cogsci.nl/3.1/manual/opensesamerun/

If you search the forum for this, you will also find plenty of examples of how you can use opensesamerun.

Cheers!
Sebastiaan

TomArmstrong

Thanks, Sebastiaan. The shuffle_horz advanced loop operation would definitely take care of the issues most of our thesis students might encounter. And while it involves writing a line of code, it is probably much simpler than creating a nested loop in E-Prime. By the way, the loop preview function is really useful!

I was recently looking for something like nested loops in OpenSesame because I wanted a mixed factorial design--I think that is what you would call it. In the study, participants see two pairs of images. One pair is a disgusting image accompanied by a neutral image. The other pair is a threatening image accompanied by a different neutral image. One pair is presented several times in a block, then the other pair is presented several times in a block. I have a block-level loop that randomizes the order of the blocks (which image pair they see first), but I also wanted to randomize the assignment of the accompanying neutral image, such that half of participants will see the disgusting image with neutral image 1 and the threatening image with neutral image 2, and vice versa. I was able to use inline script objects to achieve this, but I was wondering if there was a way to do it with advanced loop operations, instead.

Also, I appreciate the code for speeding up experiments!

DahmSF

@sebastiaan:
I have searched around in the forum. But i could not find a post on my issue.
When i understood it correctly, opensesamerun is used via the command line. This is not helpful, if participants have to start the experiment on their own. You know, participants can do strange things. And not all of them know how to use the command line. Anyway, i would need to tell each participant their individual path of the experiment (they use their own laptops). Hence, it would be great to save a shortcut, which opens the experiment by doubleclicking.
What i have in mind is a further option in "save as" to save the shortcut of the experiment directly.

sebastiaan

@DahmSF While searching for a link to an example batch file, I noticed that you asked the same question already here , where @Jarik gave you a perfectly reasonable answer.

Please help us maintain a clean forum, and don't double-post questions!

DahmSF

Just an idea for the forum. In some occasions scientists really have time pressure ;-) and it would be nice to to know whether someone is already working on an answer to your question. Hence i thought whether it is possible to mark a comment with "I am working on an answer". Something similar like the "thank" button, which is nice.

sebastiaan

Hence i thought whether it is possible to mark a comment with "I am working on an answer".

Yes, that's a good point. It would also avoid crossposts when multiple people are replying to the same discussion.