Images for Trials

megan0802

Hi,

I'm currently using open sesame to create my dissertation task. I have 2 tasks. My encoding task has 40 trials (each trial has an object and a face) but I have 4 categories:
Positive face and criminal objects (10 trials)
Positive face and neutral objects (10 trials)
Negative face and criminal objects (10 trials)
Negative face and neutral objects (10 trials)
So each trial needs to have different images but I'm unsure how to code this to tell open sesame to have different objects and faces per trial.

Then in my recall task I have 80 objects (40 new and 40 old) and participants are asked 'old or new?' So again I'm unsure how to 'tell' open sesame to use different objects per trial?

eduard

Hi Megan,

You should follow the Step-by-step tutorial. It covers how to use different images per trial. Once you have that, the rest should be a piece of cake.

Eduard

megan0802

I have followed that twice now but can't seem to get it to work for my experiment as it's not just 2 images (so faces and objects) I have positive and negative faces and criminal and neutral objects so don't know how to 'tell' open sesame to choose a specific type of face or object.

If it could be any face or object I know I could just put [face] or [object] in the script but because I have 2 types of faces and 2 types of objects I'm unsure how to do this?

eduard

How about you have something like [face_type1],[face_type2],[object_type1],[object_type2], or probably easier: [image]. In your loop_table, you then define what image refers to in each condition. See example experiment attached. (Note, it won't work, as I have not added any images to the file pool).

Eduard

megan0802

Hi Eduard,

Attached is what I currently have on my loop file, but don't know whether this is correct or if i should rename them to what you've said? The labels i've used are the actual names of the images in my file pool

eduard

If faces and objects should be shown in the same trials and either one or the other, I strongly recommend having only one column with all the image names. To classify the images to which condition they belong, you can add more columns, e.g. image_type with values face or object, or two separate columns (face - yes/no, object - yes/no). The variable names that you give them are (almost) completely arbitrary.

Oh, and for the image names, you should just provide the file extension to the images. This will make showing them easier (e.g instead of "ax", "ax.jpg", or whatever your file format is)

Eduard

megan0802

for each trial i will be showing both a face and an object, so surely if I just had one column each trial would only show one or the other depending on whether i had written 'face' or 'object' in that row? Or have I got that wrong? Ok so in terms of images I wouldn't have just the name of the object, instead for example it would be 'axe.JPEG'

Sorry for all the questions, I'm just unsure how to have different images for each trial due to their being different face and object types.

eduard

for each trial i will be showing both a face and an object, so surely if I just had one column each trial would only show one or the other depending on whether i had written 'face' or 'object' in that row?

Yes. In that case you need two columns

Or have I got that wrong?

Nope.

Ok so in terms of images I wouldn't have just the name of the object, instead for example it would be 'axe.JPEG'

Yes.

Sorry for all the questions

No problem, that's what we're here for (well, maybe not at that time)

megan0802

ok so haven't I already got the 2 columns? As I have faces then all the 40 trials for faces listed then objects with all the 40 objects? If that's correct would you then just say I need to add another column of 'image' where i list all the objects as what they would be in my file format? So 3 columns altogether? Faces, objects, image?

I've just edited my task on open sesame, so would it be something like whats attached?
Yes thank you, just trying to sort it out asap!

eduard

Yes, you have the two columns. Earlier I didn't know what exactly you wanted. You don't need a third columns for the images, per se. You can do that, but that's a little redundant. Just add the the file extension to the face and the object columns, and you have all you need. If you like you can add two more columns. One for facetype (positive, vs negative), and one for objecttype (neutral vs. negative), or whatever your categories are, so that your analysis will be easier later on.

You see?

megan0802

ok great thank you! This is going to sound a silly question but how do I add the file extension to the face and object columns I already have? Would it just be adding 'JPG' or 'JPEG' after each of the words i've already got listed?

Yes that will probably make sense, as I need to know what types of objects and faces have been in each trial.

eduard

Double click in the field, and baam, you're in editing mode.

Btw. There are advanced loop options. I am not very familiar with them, but if you have your list properly written in some textfile you can read your loop table from that file, or even copy paste. I've never used it though, so I can't really help there.

This link might be helpful
http://osdoc.cogsci.nl/3.1/manual/structure/loop/#reading-independent-variables-from-file

megan0802

Great thank you! So you are suggesting me to rename all my for example 'postive1' and 'axe' to 'positive1.JPEG' and 'axe.JPG'? Sorry just want to clarify this is what you meant.

Ok i'll have a look at that - what would I use that for?

Also, attached is my recall trial. Would you suggest that I again rename my objects with the file format, so 'axe.JPG' and this would the be okay?

eduard

To not have to type "JPEG" 40 times.

But yes that is exactly what I am suggesting.

Good luck!

megan0802

Thank you for your help! Hopefully I can get this working now!

megan0802

Since doing everything above, my experiment now actually runs as in a trial of a face and an object flashes up, followed by a fixer dot. However, when moving onto the next trial it uses exactly the same images as the first trial. From reading examples, I think this is happening because in the 'script' I haven't put my variables in square brackets (which I think is what would tell open sesame to change the picture within that variable per trial?) Since I have 2 types of faces and 2 types of objects i'm not sure where in the script I would put this information and also what I would actually write in the square brackets, as simply putting [faces] or [objects] won't be enough as i have 2 different types within each.

asthanakm

Hi there,

I am a new user of open sesame and I am also having the similar problem and the error I am receiving is the following:

********item-stack: ``
exception type: OSError
exception message: [Errno 2] No such file or directory********

I am sending my experiment file. Please have a look and let me know where I am wrong. Your help regarding this endeavour will be appreciated by me.

best,

Manish

eduard

Hi Megan,

As long as you have only two columns in your loop table (faces and objects), the actual type of face of object, does not matter for the logic of the presentation. You have to see it like that: Opensesame does not care about the content of the images you present. It only cares about the location of the image (the correct path) and the position in the experiment, where and when, to present the image. So, your experimental manipulation won't change image presentation.
For example, in the beginner's tutorial, you just have a single command for the presentation of the cue, even though there are actually three types of images (neutral, left, right, I believe). Do you see what I mean?

If you just use [ "VariableNameForTheFaces" ] and [ "VariableNameForTheObjects" ], it will present line by line what was defined in the loop table in separate trials.

Eduard

eduard

HI Manish,

You have to make sure that the images that you want to present are either in the filepool, or at least in the same folder as your experimental file. The error message you get is actually quite accurate, Opensesame can't find the file you want to present. If you call [Response] and [stimuli] in your sketchpad, these placeholders are replaced with whatever value are listed in those columns. In your [Response] column, there are no image files (no file extensions), and in your [stimuli] column many images that you want to show are not in the file pool.

Does this make sense?

Eduard

megan0802

Hi Eduard,

I just have two columns: faces and objects, with all my trials then listed. However, when writing in the script '[faces]' and [objects] it's telling me that these variables aren't recognised, so when i go to run the trials they just use the same images.

Attached is what my loop table looks like, does this look correct? Also, attached is the script of the loop table and the sketch pad, as i'm not sure where i would write my variables in the square brackets to tell open sesame to change this variable per trial.

eduard

Your variables are called [Faces] and [Objects]. Are you sure you haven't made a typo?

megan0802

Yes that's what i wrote in the brackets. is there a specifc place i need to write those 2 variables in the script? Where it says 'draw image center' i've tried deleting what it says there 'positive1' and 'axe' and replacing those with [faces] and [objects] but it says the variables aren't recognised.

eduard

Please confirm that replacing [faces] and [objects] with [Faces] and [Objects] does not help.

megan0802

Yes this doesn't work. Where would I actually write [Faces] and [Objects] in the script? Is there a specific place or anything I have to delete in order to get this to work?

I've tried multiple times but it just says the variable isn't recognised!
This is what my script shows, so where would I write my variables?

eduard

I don't want to sound mean, but if you have done the beginner tutorial twice, you should know that. Step 6 explains exactly your problem. If you want to update your images every trial, you have to change the parameter that defines which image will be shown, which is file. And instead of showing the same image on every trial (as happens right now because you hardcoded positive1.jpg and apron1.JPEG), you use the variable name, that includes all the files.

So, the be very clear. replace positive1.jpg with [Faces] and apron1.JPEG with [Objects].

Alright?

megan0802

Yes completely understand what you're saying and what you've said for me to swap i've already done as I thought this was the problem. However despite only having columns of Faces and Objects, when writing [Faces] and [Objects] in the script it still says the variable isn't recognised.

eduard

Okay, then please upload your experiment. I'd like to have a look

megan0802

Attached is my experiment. The variables on there are still 'positive1' and 'apron' as when i've tried to change it it says the variables don't exist and doesn't save it.

eduard

You have to be accurate when you read the error message. Opensesame didn't complain about non-existent variable names (both [Faces] as well as [Object] does work!), instead it was not finding the files to present. This is because the files in your file pool and the files in your loop table do not correspond, you have many typos in that list (e.g. underscores instead of spaces, Uppercase instead of lowercase, etc.).

You'll have to clean up everything. So, make sure that the exact same name is used for both file as well as entry in the list. Furthermore, in order to test whether everything is good, you should also change the duration of the image sketchpad. Currently it is set to 0, which effectively means that Opensesame is skipping it (because there is a fixation dot, presented after).
Also, finding all the problematic strings will be easier, if you do not present the images randomized, but sequentially (settings in the loop table).

Good luck fixing it.

Eduard

megan0802

Finally managed to get it to work! All 40 of my trials play followed by a 'break' screen and then 'instructions' which are for my second task. However, instead of going onto my second task it just repeats my first? When looking in the 'experiment - sequence' tab 'task 2 loop' is listed but not 'block loop 2' which is where i have listed all my trials for my second task. I'm assuming this is why this task won't work, but I'm unsure how to change this?

eduard

Hi,

Yes, this sounds like the experimental sequence is a little messed up. Basically, Sequences are repeated as often as indicated in the loop table. Only after that it will proceed with the next item in the next outer sequence. The indentation in the overview area should give you a good idea of whether order in which things are executed and which item is part of which sequence.

Without seeing the sequence myself, I can't say anything more specific to help you. Uploading the experiment would help .

Eduard