[open] Stimulus presentation, keyboard timing and feedback beep

Comments

• eduard

  March 2016 edited March 2016

  Hi Rhett,

  How about a version 4

  
  # Basically, it assumes  that you have your three displays already prepared in 
  # a canvas. Then, in a while loop, you keep on presenting a certain display/mask
  # depending on the current time. So that after 300 ms/3000ms, etc, the display
  # is updated
  
  # first, you have already prepared your three stimuli and your mask
  all_displays = [stim1_cv,mask1_cv,stim2_cv,mask2_cv,stim3_cv, mask3_cv]
  
  # then it is also handy yo have a list with the durations of each display
  all_timeouts = [300,3000,300,3000,300,3000]
  
  # create a keyboard to poll for the responses
  kb = keyboard(keylist=['SHIFT']
  
  # also it should be known which key corresponds to the correct response
  correct_key = 'z' # just an example
  
  # give response times and correct response default values
  var.resp_time = None
  var.correct = None
  
  # finally, I assume that you already set up the sound file
  sound = sampler(file, volume=.5)
  
  # sample time, when the trial begins
  init_time = clock.time()
  
  # loop over each display
  for i,cv in enumerate(all_displays):
      # show one display
      t0 = cv.show()
      # poll for responses (waiting for respective timeout (300/3000ms))
      key,time  = kb.get_key(timeout=all_timeouts[i]) 
      if key != None:
          sound.play()
          # set whether response was correct
          if key == correct_key:
              var.correct = 1
          else:
              var.correct = 0   
          # I assume the very beginning of the trial (before the very first display)
          var.resp_time = time-init_time
          # if a key was pressed, you want to break out of loop
          break
  

  I hope this code makes sense to you. It doesn't work as it is right now, because I don't have defined any canvases and also no sound. If you need help, just ask or check out the documentation on how to do these kinda things in inline_scripts: http://osdoc.cogsci.nl/python/ (in particular, check out canvas and sampler)

  Good luck,

  Eduard

  Edit: To log the variables, you either have to use log.write_vars() in the end of your code, or add a logger item after the inline_script.

  
• rhettclayton

  March 2016 edited 5:27AM

  Hey Eduard!

  I think I am beginning to understand your code. Thank you very much for your response!

  The actual stimulus sketchpad consists of three images displayed along the x axis. Their filenames are variables that are randomly changed within the experiment loop.

  When I run your code, with the canvas defined in the prepare stage, it says that the variables that are the file names (i.e. one image points to [Center_Object].jpg, and [Center_Object].jpg has different values (i.e. Face_1, Face_1, etc) between each set) are not defined.

  How do I prepare sketchpads in advance of running that script, where objects within the sketchpad are linked to variables?

  Thanks!!

  Rhett
• eduard

  March 2016 edited 5:27AM

  Hi rhett,

  If you want to draw 3 images on one canvas, you need to provide all the names for each file. One way to get this done rather easily, is to have three variables in the loop table, e.g. file1, file2,file3. Later in the inline_script, you then have to initialize and draw the images on it.

  # init canvas
  display = canvas()
  
  # draw images on the canvas
  display.image(file1,x=-300,y=0,scale=0.5)
  display.image(file2,x=0,y=0,scale=0.5)
  display.image(file3,x=300,y=0,scale=0.5)
  

  Of course, you need to make sure that the filename is a valid path to the files.

  Is this what you wanted to know? Let me know if you need more info.

  Good luck,

  Eduard

  
• rhettclayton

  March 2016 edited 5:27AM

  Hey Eduard,

  I actually have the code in place for the images, but in the 'loop' object, there are variables (center object, left object, right object) that change between sets (i.e. center object becomes t_n1.jpg, left object becomes t_t1.jpg, etc)

  In the code, it looks like this http://imgur.com/a/LhKOc

  Hope this helps to elucidate my question!

  Thanks!

  Rhett
• eduard

  March 2016 edited 5:27AM

  Hi Rhett,

  Sorry for the late response. Unfortunately, I'm still unsure whether I understand your issue. Let's take a step back and make sure we're on the same page.

  there are variables (center object, left object, right object) that change between sets

  With "change between sets", do you only mean, that the file names are different between trials of your experiment, or in some other way?

  According to how I understand your design, the only thing that changes between trials is which images are presented. Everything else (position,timing, response sampling) doesn't change, right? If so, which parts of the design are already working as you want them to?

  I actually have the code in place for the images

  Nice. But again, according to my understanding, having the image presenting part set, goes hand in hand with having the loop structure set. If this is not the case for you, maybe you should consider rethinking the image presentation code?

  I hope this make sense. Let me know whether I get you right.

  Good luck,

  edaurd

  
• rhettclayton

  April 2016 edited 5:27AM

  Hey Eduard,

  Thanks for your patience.

  I'll step it back~

  In the 'prepare' phase of my script, I have the stimuli and mask sketchpads written up (as per the screenshot in my comment above)

  The path of the file is path = exp.pool[[Flank_Object].bmp], 'Flank_Object' is a variable that changes between trials, from t_t, to t_t2 etc. The files in the file pool are named t_t1.bmp, t_t2.bmp etc

  When I run the OpenSesame task, it errors out when it comes to running the script part (i.e. it runs everythins before that) saying:

  item-stack: MainRun[run].Tr_Face[run].Tr_Face_Seq[prepare].Tr_Face_Script[prepare]
  exception type: NameError
  exception message: name 'Flank_Object' is not defined
  item: Tr_Face_Script
  time: Fri Apr 01 10:38:22 2016
  phase: prepare

  Any idea as to what might be causing this?

  Here's a link to the experiment at this point in time~ http://bit.ly/1Y2SjFp

  Thanks!

  Rhett
• eduard

  April 2016 edited April 2016

  Hi Rhett,

  Here a version of that prepare phase that doesn't throw an error:

  Tr_Face_Flank_Image = canvas()
  path = exp.pool[str(var.Flank_Object)+'.BMP']
  Tr_Face_Flank_Image.image(path, center=True, x=480, y=None, scale=None)
  path = exp.pool[str(var.Flank_Object)+'.BMP']
  print path
  Tr_Face_Flank_Image.image(path, center=True, x=-480, y=None, scale=None)
  path = exp.pool[str(var.Central_Object)+'.BMP']
  Tr_Face_Flank_Image.image(path, center=True, x=0, y=None, scale=None)
  Tr_Face_Flank_Image.show()
  clock.sleep(500)
  Tr_Face_Visual_Mask = canvas()
  Tr_Face_Visual_Mask.show()
  

  A couple of comments:

  • once you created an instance of a canvas (let's say, you called it my_cv), you have to use this label to call all the canvas functions. That is use my_cv.image() instead of canvas.image

  • It might be easier and more efficient, if you're loop table already contained the path towards the files, not just the name, so also including the suffix (instead of t_n1, use t_n1.BMP as value in the table). Once you do it that way, you don't have to go through the trouble of adding the extension to the file (the main reason why it didn't work for you)

  • when you want to access variables that you define in the loop-table from within an inline_script, it is not [variable] (like it would in sketchpads and such), but rather var.variable.

  I hope this makes more sense now.

  Eduard

  
• rhettclayton

  April 2016 edited 5:27AM

  Hey Eduard,

  Thanks for your help! I think that resolves that problem!

  However, some other ones have come up.

  Firstly, the fixation cross only appears after the first set of facial images, and not before. I can't decipher what is telling the faces to appear prior to the fixation dot?

  Secondly, past the fixation dot / facial mixup, the experiment errors out, claiming that response_time is not defined. I have not had to define it in my other loops (i.e. the ones I have not implemented your code in) and it records time accurately in those loops. What am I missing?

  Thanks again for your help!

  Rhett
• eduard

  April 2016 edited 5:27AM

  Hi Rhett,

  Firstly, the fixation cross only appears after the first set of facial images, and not before. I can't decipher what is telling the faces to appear prior to the fixation dot?

  Are you sure that the fixation cross is really not there? Sometimes, it just appears so, but in fact Opensesame proceeds to the next item super quickly, so that you just can't see it. In this case, changing the duration feature would fix it. Maybe it is also a problem of the ordering of the items in your experiment? Obviously, things that appear first in a sequence are executed first. I guess these are the most likely reasons for the fixation cross to not appear.

  past the fixation dot / facial mixup, the experiment errors out, claiming that response_time is not defined

  I'll have a look on it tomorrow, and let you know what could have caused that.

  Eduard

  
• eduard

  April 2016 edited 5:27AM

  Hey Rhett,

  the script in TR_Face_Script has code in the prepare and in the run phase. Of all elements in a sequence Opensesame executes first all prepare phases and moves then on to the run phase. This probably explains your first problem, right?

  Indeed, the variable response_time is not defined, still you want to log it. The variable that logs the response time in your first sequence is called var.resp_time. So you would have to use this variable name to log it. However, variables with the prefix var are automatically logged. So you don't need to add it manually.

  Does this help?

  Eduard