Negative time values in Log file & out-of-range responses recorded

In looking through a new data set, I came across two instances where a stimulus was presented, a 10 was recorded as the response from the DirectRT button box (in the input file, time was rt:1,9). These two trials also had no time value recorded in the output, but a negative time value was recorded in the Log file (-518547 in one case, -677371 in the other). It only happened on two out of more than 11,000 trials so it's not a major catastrophe...but in one case it ended the experiment early (the third block of trials did not appear for the participant). In any event, I'd like to understand what is going on here.

Some additional information - the errors occurred with different participants, collected on different computers, on different days/times, and with different stimuli.

Have others had this happen? Does anybody know the reason?

Thanks...
David

What version of DirectRT (200x.x.x) are you running and can you post a sample input file and corresponding log version of its resulting data file?

[quote=jarvis24;3709]What version of DirectRT (200x.x.x) are you running and can you post a sample input file and corresponding log version of its resulting data file?[/quote]

I'm running DirectRT 2008.1.0.11. A sample format and input line follows:
[INDENT]block,trial,!prime-type,bgr,wgr,style,stim,loc,time,stim,loc,time,stim,loc,time,stim,loc,time,*[/INDENT][INDENT]1,71,NONWORD,0,1,1,~Press any button when ready.,"0,0,1",rt:any,~,"0,0,1",200,~+++,"0,0,1",500,~apsciponde,"0,0,1","rt:1,9,clear500",*[/INDENT]
The log file output looks like this (I'm pasting output from two trials here - one that worked and one that didn't):
[INDENT]Date,Time,File,Subj,Cond,Block,Trial,Order,prime-type,Resp,Name,Stim,RT,Correct,Started@,LoadTime,TotalInt,Stim 1,Time 1,Stim 2,Time 2,Stim 3,Time 3,Stim 4,Time 4


4/10/2008,10:26,R:\Active Projects\Motivation & Info Processing\Lexical Decision Task Study\lexical decision task (sp08-1).csv,208,1,1,1,185.01,NONWORD,8,8,Press any button when ready.,523,TRUE,659365,26,1012,Press any button when ready.,547,,212,+++,507,xemniest,614


4/10/2008,10:26,R:\Active Projects\Motivation & Info Processing\Lexical Decision Task Study\lexical decision task (sp08-1).csv,208,1,1,1,185.02,NONWORD,1,1,xemniest,611,TRUE,659365,26,1012,Press any button when ready.,547,,212,+++,507,xemniest,614


4/10/2008,10:26,R:\Active Projects\Motivation & Info Processing\Lexical Decision Task Study\lexical decision task (sp08-1).csv,208,1,1,2,190.01,NONWORD,8,8,Press any button when ready.,503,TRUE,673624,26,1012,Press any button when ready.,520,,213,+++,507,aopenlktcm,-677371


4/10/2008,10:26,R:\Active Projects\Motivation & Info Processing\Lexical Decision Task Study\lexical decision task (sp08-1).csv,208,1,1,2,190.02,NONWORD,0,,,0,FALSE,673624,26,1012,Press any button when ready.,520,,213,+++,507,aopenlktcm,-677371[/INDENT]Another unusual feature that I'm noticing is that the log file seems to list 8 and 2 as response values (instead of the 9 and 1 that show up in the abbreviated output file) - strange.

Thanks in advance for any thoughts you can offer...

I apologize for the hassle but could you post the actual files as attachments?

No hassle at all - DirectRT input and Log output files for one of the problem cases are attached (error occurs on lines 252-253 of the log file). Thanks for taking the time to look at them...

Is it just a coincidence that it happened on the final trial in the sample you attached or is that the norm? i.e., if you sort by the order column. Also, if you use the keyboard rather than the button box, does that affect the 8 vs 9 output in the data file?