[solved] Structuring experiment with loops and sequences for randomisation

Comments

• JoshuaBaker

  February 2016 edited February 2016

  Hi Sofiaa,

  you can create your trial list in a loop item. Just create some variables with a number of 'cycles' (i.e, trials). then in the top right of the loop item is an option to run sequentially or randomly. choose random! the loop will then run your sequence that you have made, x number of times until all of your trials are been presented once (assuming you have not set a repeat). make sure in your loop item that you choose your trial sequence in the 'item to run' field. given the two items stay the same on every trial, you could either have lots of repeats in your trial list (in loop item), or you could insert them into the sketchpad. To only randomise the picture, add in 'any' picture into the sketchpad, then double click on it with the selector (arrow) tool, and edit the line of code from something like file="F016.jpg", to file="[pic_variable]". Make sure it exactly matches the name of the variable containing the picture names in the loop item.
• sofiaa

  February 2016 edited 2:39PM

  Hi JoshuaBaker,

  first of all thank you so so much for your help. I already understood more than at the beginning :-)
  I have created a variable in the loop_item (named picture) and 36 cycles in random order. In the 36 rows of the column I have written downwards the name of the pictures (picture1, picture2 etc).
  Where do I include/upload the pictures now? Do I include all 36 pictures/ sketchpad_items in the trial_sequence? Do there have to be 36 sketchpad items in the trial_sequence? Where do I have to add them in the experiment structure?
  I hope it is understandable what my question is about :-)
• JoshuaBaker

  February 2016 edited 2:39PM

  no no, no need for 36 sketchpads! just the one sketch pad. open the sketchpad on the left hand side there are a number of icons (one being a picture insert). click that, and select the images from your filepool. You need to add all stimuli into your filepool first. then do as i suggest in my comment above, use the pic_name variable in square brackets in the line of code in the image properties in the sketchpad item (double click on the image and edit the code). Everytime that the sequences gets to the sketchpad item, the value from the variable pic_name will be used from your loop table. It might be trial number 19 for example, and so pic_19 will be presented on this occasion.
• sofiaa

  February 2016 edited 2:39PM

  After many attemps yesterday and today it works now! Thank you so much !! Now I think I have understood how loops, sequences and randomisation work. :-)
  I have got one last question. I first used a different program where I could not reconstruct the randomisation of my pictures so thererefore my data was basically useless. Can I see in my data table to which picture the values of my two items were assigned? Will the data output show me for example: picture 19 --> item 1 (value 2) --> item 2 (value 7); picture 23 --> item 1 (value 1) --> item 2 (value 4) etc... I ran the experiment to test it but the excel table is still quite confusing for me.
• JoshuaBaker

  February 2016 edited 2:39PM

  of course.

  you just need to select a bunch of variables to log in the logger item. untick the 'log all variables', and manually type in a few are going to be useful. type them exactly as they appear in the loop trial table. so, picture, item1, item2, response_keyboard_response
• sofiaa

  February 2016 edited 2:39PM

  So if I let the option "log all variables" activated I can't reconstruct the randomisation? I'm a bit insecure in adding variables on my own because it is not recommended.
• JoshuaBaker

  February 2016 edited 2:39PM

  you can still randomise your trials and see in what order they appeared. that's what a logger can do. you can make sure that all the information you need in a trial (e.g. what picture it was, what colour it was, what size it was, what condition (e.g. face or non-face, male or female)).. then when you look at the log file, you can see that trials were presented in a random order, and you can see what order they were given in by looking at the data in the variables that you have logged. If you really wanted to, create a variable in your loop item called trial_number, and just assign all of your trials a number(e.g. 1 2 3 4 5 6 7 ...). if you present these randomly, look in your log file at the variable trial number, and you can see what order they were presented.
• JoshuaBaker

  February 2016 edited 2:39PM

  feel free to email me so that i can assist you further (I also have a guide I have made)

  joshua.baker@ntu.ac.uk
• sofiaa

  February 2016 edited 2:39PM

  Thank you so so much for your help and also for your offer of further assistance. I'm going to try it out with the variable "trial_number", sounds like a good and safe option.
  Great work JoshuaBaker! You helped me a lot!
• JoshuaBaker

  February 2016 edited 2:39PM

  my pleasure!