Preference task

SuperLab SuperLab 4 Support
1

I’m working on an experiment trying to create preferences for neutral stimuli. The task I’ve created has participants look at random pairs of stimuli and press a button to indicate which they prefer. The task itself is all set up in Superlab, but I’m having a hard time figuring out how to extract the data without having to look at each participant’s responses and tallying them up. Is there a way I can set the program up so that I’ll be able to see what percentage of the time participants chose A, B or C? or can I only measure correct vs. incorrect responses?

2

Currently this is something you will have to do outside of SuperLab with a program such as Excel or another statistical package.

3

Preference test using image pairs

Hi Dennis and Monika,

I joined this conversation because I think my experiment might be similar.

I am creating a visual preference test where participants see a number of image pairs which must be presented together at the same time. Each image pair is presented for 5 seconds and then disappears, at which time, while looking at a blank screen, the participant picks which of the two images they preferred (image on right or image on left). I thought maybe the set-up that you created might help me, Dennis.

On a different note, the order in which the image pairs are presented must be randomized for each participant (which I tried to do on the block level), but additionally, the side in which images are presented within an image pair should also be randomized (right and left side) for each participant. I am unsure how to do this part, and also confused about how data will be collected as within an image pair, sometimes the image on the right is “correct” and sometimes the image on the left is “correct”.

Thank you for any feedback!

4

Update with Preference test

Hi there,

I have been working on the experiment, and what I have so far is attached. I am trying to make the BLOCK Stimuli randomize the order in which it plays TRIAL Stimuli 1 or TRIAL Stimuli 2. Additionally, in the EVENTS section, their are two stimuli found under Eyes and two stimuli under Ears. When EVENT Eyes plays, I want it to randomize which of the two eye stimuli is shown to the participant, and I do not want both of the Eye stimuli to be shown, just one of the two. The same thing goes for Ears.

Additionally, I want people to make a decision regarding which image they prefer not while the images are up, but when the images disappear and participants are looking at a white screen. I’m not sure how to do this.

Thanks for your help!

Dogs2.zip (1.68 MB)

5

If your experiment is more extensive and has more stimuli, I would suggest using Trial Variables for Screen Location for the position of your pictures. This is found under the File menu. Below are the steps that were used to create the updated experiment that is attached.

  • Create a code. In your experiment it is named "Presented Eyes". This code has two values "Yes" and "No" 
   <ul>
  • You will also create a code fro "Presented Ears" with a "Yes" and "No" value
    • In the "Eyes" event a [B]Feedback[/B] action was added. You will present Feedback: Always at the end of the event and Change value of code "Presented Eyes" to "Yes
    • The same type of Feedback was added to "Ears" except you use the code value "Presented Ears"
    • In the [B]Trial Editor[/B] for "Stimuli Eyes", under the [B]When to Present[/B] tab, an expression was created to read "Current Trial's Code "Stimuli Eyes" Is Not Set To "Yes"
    • The same thing is done under the [B]Trial Editor[/B] for "Stimuli Ears"
    • Two macros were created, "See if we presented eyes" and "See if we have presented ears" This macro is set to run After each trial is presented.
    • Under the [B]Expression [/B]tab it states "Number of Trials Presented in This Block is equal to 1 trials (Using only the last 1 trials in the block and Trials with code "Presented Eyes" equal to Yes")
    • The [B]If True[/B] tab states "Change value of code "Presented Eyes" to "Yes" in all trials in this block
    • The same type of macro was created for "See if we have presented ears"

    Dogs2revB.zip (1.68 MB)

    • 6

    Thank you!

    I got caught up in another project and will be coming back to this study in the next week. Thank you very much for your help, and I might have a few additional questions.

    Cheers!

    7

    Continuing the project

    Thanks for your help! Your comments were very helpful, and I’m now trying to craft the experiment into its real form, and I’m having some trouble.

    The overarching issue is that I am trying to figure out how to simplify the experimental design in SuperLab. You recommended using Trial Variables for Screen Location, but I am unsure how to program this to meet my experimental needs.

    I will be presenting about 90 different image pairs, and while the order of the image pair presentation will be randomized, the actual images that appear together within a pair will not be random. At the same time, I do want to randomize the side of the screen on which the images appear within each image pair. To accomplish this, I presently have it set up so that in the stimuli list, I give SuperLab the option to present Dog 1_A_B or Dog 1_B_A for one particular image pair. A is always bigger and so I tell the computer to present one of the two image pairs, but not both.

    I am still not sure the best way to create the stimuli lists. You’ll see I created a stimulus list for J_Dog, but I don’t think that is the correct way to do it as I don’t want both derivations of each image pari presented.

    Is there an easier way to create a stimulus list and I am just missing something? I have created each image as an individual JPG, and I am not sure if there is an easier way to create the stimulus list and randomize the side presentation within an image pair.

    I also noticed that becase my participants make their choice when the stimuli have disappeared and participants are looking at a screen prompting them to make a choice, the participant response is not in line with the image in question. Is there any way to change that?

    Thanks for any guidance you could provide. I’m also available by phone if that’s a possibility.

    Cheers!

    Dogs2revB.sl4 (11.3 KB)

    8

    This is something that will be faster if discussed over the phone. I will contact you via private message to set up a time.

    9

    Thank you!!!

    My experiment looks great! Everyone at SuperLab is incredibly helpful. I used to feel like this :frowning: but now I feel like this :slight_smile: :slight_smile:

    Thank you!!!