presenting a 3x3 grid of (random but different) letters


I am trying to design an experiment where a 3x3 grid of letters is presented to the participant (9 letters). I want the computer to chose 9 random but, crucially, different letters (consonants) on every trial. I have created 9 stimulus lists each consisting of the same 21 consonants). I don’t want a repeat of the same letter within a trial. After the grid of letters is shown for 1000 ms, the participant should respond.

I have 2 problems that I’m hoping you can help me with:

  1. I do not know how to have the computer randomly select different letters from each of the stimulus lists. I have randomized in the trial editor by choosing the “unpaired” option, so that any letter from any of the 9 lists is chosen on every trial. However, my problem is that the letters often get repeated within the same grid (e.g. DD)

  2. I want the 3x3 grid of letters to be shown ONCE only on any given trial, and then for a response from the subject. However, as I now have it programmed, the grid of letters is being shown multiple times. I am not exactly sure why (although I have a hunch it is because of the randomization in the trial editor that I have selected) or how to correct this.

I very much appreciate any help you can give. I am attaching a copy of the experiment to this message.


IM 2015 superlab 5 (3.37 KB)

Hi Jason,

Here are the answers to your questions. Also, I attached a sample experiment to show the basics of getting a grid-style experiment properly structured in SuperLab.

I will answer your questions in reverse order.

Answer to Q2: To understand why the grid is happening multiple times before a response, use a special menu item in SuperLab named “Expand Events That Use Lists…”

  • Open the "[B]IM 2015 superlab 5 version[/B]" experiment you had attached here previously.
  • Delete trials 2 through 10. (delete them because they are not used, and they would be a distraction in the next step.)
  • Go to [main menu] >> Experiment >> Expand Events That Use Lists. (Choose that menu item.)
A new experiment will open. The new experiment shows that your original trial "1" will be expanded to 21 trials whenever SuperLab runs the experiment.

The reason there will be 21 trials is that your list(s) contain 21 items.

Viewing experiments with “Expand Events That Use Lists” is intended to illustrate how SuperLab processes a list when you run the experiment. Making some very small “throwaway” experiments and then expanding them can be a good way to train your intuition about what kind of behavior to expect from a list-based experiment.

Answer to Q1: After you have expanded the experiment (as described above), you can also use the expanded version to answer question 1. The reason why you can end up with DD (two D letters) in the same grid is that each grid position is created using a unique event. Each event draws a letter from the pool of 21 letters, but each event has its own pool of 21 letters.

Please refer to the attached demo experiment. This experiment has only 1 stimulus list. In general, there is almost never any reason to create several identical lists. Since you only need the same 21 letters, you only need 1 list containing the letters.

In this new sample (see attached), there is only 1 trial named “grid”, and one event named “grid”. Open the “grid” trial and look at the tab named “When to Present.” You can see there is a conditional expression that will cause this list-trial to stop presenting trials past a certain number. That number is 10. (The reason it counts out to 10 instead of 9: i am counting the “instructions” trial, too. “Instructions” plus the 9 letters adds up to 10 before we want to stop.)

The other important feature being used in this sample is Trial Levels. You can see these levels (which i named “positions”) by going to [main menu] >> Experiment >> Trial Levels. The “grid” event uses these position levels to lay out the letters in the 9 different spots of the grid.

grid_demo_stimulus_list.sl5 (5.82 KB)