Not recording taps after first trial

There are two issues with the recording of taps. They are very likely related, 
however they are being entered independently for logging/tracking purposes.  

There is an issue with recording taps when using OS 10.9.2 (and above), which 
we suspect is related to issues with opening the MIDI channel.  Taps are not 
recorded starting in the 2nd trial (see attached files documenting this issue). 
 However, we also found that simply testing the MIDI tapping before the start 
of the experiment triggered a similar problem.  When hitting the “List and 
Test Devices” button and opening the “MIDI Devices and Testing” window, 
and then hitting “Test Tap Recording” the program fails to record taps in 
the experiment even on the first trial.  This happens regardless of whether we 
tap during the “Test Tap Recording” exercise.  Consequently we suspect this 
is an issue with opening the MIDI channel, rather than tapping behavior on the 
first trial.  

Some other important bits of information:
1)  We do not see this problem when using the space bar in instead of an 
external MIDI device.  However, we have replicated this issue with both a piano 
keyboard and drum pad.    Similarly, the spacebar continues to work even when 
on the second trial (i.e. external MIDI information fails to be recorded).  
2)  This is unique to OS 10.9.X.  We tested the same build (222) on various 
10.6.8 and 10.8.3 systems and did not see the problem.  Similarly, we also 
tested several past builds (i.e. 178, 159) on 10.9.2 and found the same problem.
3)  Even if we do not tap on the first trial, the program does not record 
tapping starting on the 2nd trial.  So it appears perhaps just opening the MIDI 
channel at the outset of the experiment creates the problem.
4)  This is independent of whether the program is in the warmup trial or the 
main experiment. We tried omitting trials (i.e. setting numWarmupTrials=0), and 
found similar behavior. 

Note: the following information pertains to both issues, and is included in 
both bug reports

To try and narrow down the number of variables to consider, we tested a variety 
of builds on a variety of operating systems.  Builds 222, 178, 159 all display 
this problem when running 10.9 (we’ve tested 10.9.2 and 10.9.3).  And we’ve 
observed this when using either a MIDI keyboard or our (more typical) drumpad 
setup.  We have not seen this problem when using build 222 on 10.6.8 and/or 
10.8.3.  Therefore it appears the issue is one of changes to the OS in 10.9.3.

Currently, we have moved to 10.9.2 for most testing, but kept a few 
“legacy” computers running 10.6 for compatibility with the older versions 
of this software (which used Apple sounds that were not longer accessible in 
later OSs).  To streamline operations we decided a while back to pick a new 
“target” OS and then just focus on that.  I need to purchase some new 
computers soon, and it appears Apple is now up to 10.9.4!  I’m fine up 
updating the newer computers to a newer version of the OS  - obviously I just 
want to make sure the MIDI stuff works OK there.

Original issue reported on code.google.com by schutz.m...@gmail.com on 4 Jul 2014 at 7:50

Attachments:

[Case 2 - Not recording taps after first trial.zip](https://storage.googleapis.com/google-code-attachments/auditory-research-suite/issue-25/comment-0/Case 2 - Not recording taps after first trial.zip)

MAPLE-Lab / auditory-research-suite

Not recording taps after first trial #25