[Rush hour] how to log responses

Comments

• sebastiaan

  January 2017

  Hi,

  To register a response, you can use the responses.add() function (no need to do item.responses.add()) as described here:

  • http://osdoc.cogsci.nl/3.1/manual/python/responses/

  You have the response (inputchar), but not the response time:the time difference between the onset of response collection, and the time at which the response was collected. The item keyword is optional (as are all other keywords, actually): it's just to disambiguate responses in case you have multiple response-collection items.

  So basically, you need to do something like:

  rt = clock.time() - t0 # where t0 is the onset of response collection
  responses.add(response=inputchar, response_time=rt)
  

  So t0 is the moment at which response collection started. But how you can determine that is unclear from your description.

  Cheers,
  Sebastiaan

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

  January 2017

  Hi Sebastiaan,
  the response comes on the serial interface and i need to process it there. Therefore,

  File "X:\Install\OpenSesame\3.1.4-py3.5-win64\lib\site-packages\share\opensesame_plugins\syncbox\libsyncbox.py", line 231, in get_button_press
  responses.add(response=inputchar, response_time=self.experiment.time())
  NameError: name 'responses' is not defined

  Ok, this seems to work:

  if len(inputchar) > 0:
                  self.experiment.responses.add(response=inputchar, response_time=self.experiment.time())
                  print("[{}]: expected '{}', got '{}'".format(self.experiment.time(), self._syncboxResponse, inputchar))
  

  At least the debug shows:

  [14214.0]: expected 's', got 's'
  [17539.0]: expected 's', got 'd'
  [18544.0]: expected 's', got 'd'
  [18778.0]: expected 's', got 's'
  [21640.0]: expected 's', got 'a'
  [23424.0]: expected 's', got 's'

  's' is a sync trigger from the scanner, while 'a', 'b', 'c' and 'd' are responses from the hand grips.

  Just 's' is visible in the log file right now, but this will be a different story.

  I prefer to log the absolute timing of the response, where t0 is the start of the experiment. I can check the Nordic Actia logs in the lab, while right now I think the are also logging time from the beginning of the experiment.