Omission Keyboard Response Marking As Incorrect on Biopac/Acknowledge

[Megan Moran]

Hi All,

I am working on getting events in E-Prime to be marked in Biopac/AcqKnowledge via a Black Box Toolkit USB TTL device (serial port device) and we are having an issue with responses consider as an omission (or no button pressed on the keyboard) are sending signals / marking both for omission and incorrect in AcqKnowledge. The response types are correct (up arrow), incorrect (down arrow), and omission... everything else is working fine and we have checked multiple times to ensure that the hex/strings being sent are correct. It also does not seem to have a pattern of which omissions it does this for... sometimes it marks on both digital channels sometimes it only marks on the omission channel. This problem does away when we remove the event for incorrect responses on the keyboard, but this is data that we would like to have. We just simply cannot find what is causing it.

If you have ever had this issue before or something similar, please let me know how you may have fixed this issue. 

Megan

[McFarlane, David]

Megan,

First, please go to the thread at groups.google.com/g/e-prime/c/leyX9jmIJ5w and then follow the links there to other threads that discuss outputting signals/marker/triggers from E-Prime, in case anything there applies to you.

For your current situation, do I understand correctly that

- You successfully get signals from E-Prime to Biopac/AcqKnowledge under some circumstances.

- You intermittently fail to get signals for non-response trials.

Finally, the first thing I look for when troubleshooting external signals is to make sure that all output signals occur in pairs, i.e., signal followed by reset to 0, with sufficient time between signal and reset to allow the external equipment to detect it. Odd things can happen otherwise. I might also mention here that the USB trigger interface cable from BioSemi (www.biosemi.com/faq/USB%20Trigger%20interface%20cable.htm) solves this problem by automatically outputting a pulse for 8 ms, whenever possible I would recommend using this device over alternatives.

After that I would look at the details of how you output a signal in the case of no response, since I am pretty sure that you do not do that part with Task Events.

-- David McFarlane