[solved] How to ignore incorrect mouse responses?

Comments

• sebastiaan

  June 2014 edited 3:46PM

  Hi,

  Your script is a bit funky in a number of ways. First off, you define a mouse object, but never use it, apparently because you use a mouse_response item to collect mouse clicks. Second, you define target_positions list, which is walked through, but is functionally ignored. (And at any rate does not contain a list of target positions, as you would expect!)

  Below you see a modified version of your script, which should do more-or-less what you want. The idea is to use a mouse object to collect clicks in an endless loop that is broken only when a click is close enough to the target (defined by the xRed and yRed experimental variables). Does that make sense? This script is instead of a mouse_response item, so you can remove that from your trial sequence.

  It might also be a good idea to walk through a basic Python tutorial, to get a grip on using while-loops, etc.

  • http://osdoc.cogsci.nl/python/about/#learning-python

  Cheers!
  Sebastiaan

  from openexp.mouse import mouse
  from math import sqrt
  
  # Create a mouse object
  my_mouse = mouse(exp, visible=False)
  
  # The maximum error from the target center
  maxClickErr = 100
  
  # Get the coordinates of the display center, which are necessary to convert
  # from top-left to center-based coordinates.
  xc = self.get('width') / 2
  yc = self.get('height') / 2
  
  # Loop until we get a hit (i.e. a correct mouse click)
  while True:
      button, pos, timestamp = my_mouse.get_click()
      cursor_x, cursor_y = pos
      # Determine x and y error
      dx = cursor_x - xc - self.get('xRed')
      dy = cursor_y - yc - self.get('yRed')
      # Break the loop if there is a hit (use Pythagoras for click error)
      clickErr = sqrt(dx**2 + dy**2)
      if clickErr <= maxClickErr:
          break
  

  
  
  Check out SigmundAI.eu for our OpenSesame AI assistant!
  
• Darby

  July 2014 edited 3:46PM

  Thanks Sebastiaan! This works perfectly. I will definitely have a look at the python training.