[open] Keep accepting responses while experiment continues

Comments

• Edwin

  May 2014 edited 9:13PM

  Hi,

  If I understand correctly, you present a stimulus every 200 ms, but assume that participants can still respond to this 1500 ms after. This would mean you will have trouble assigning responses to stimuli afterwards.

  An example: stim1 occurs at t=0, stim2 at t=200, stim3 at t=400. If a button is pressed on t=300 and at t=500, these could be responses to stim1 and stim2 (with RTs of 300 ms), and stim3 would have been missed. Alternatively, you could have a crazy fast participant that missed stim1, but responded to stim2 and stim3 with 100 ms RTs.

  You're not the first to encounter this problem, it's quite common in dual task experiments (e.g. to measure the amount of distraction a task has on a respond-to-beeps task). The solution that seems to make most sense to me, is to record the onset of every stimulus and the times of button presses in an event log. Afterwards, you can choose an approach to bind the responses to a certain stimulus.

  Example of an event log (beeps every second):

  event,timestamp
  beep,1000
  resp, 2000
  beep, 3000
  resp, 3450
  beep, 4000
  beep, 5000
  resp, 5100
  

  It's possible to do this via inline scripting, but will require parallel processing/threading. The analysis will require some thought as well.

  Good luck!

  Edwin