Flushing serial TTL buffer

[AaronNewman]

Turns out this is a bigger problem than I initially realized. Since the button presses are treated exactly the same way, we are unable to collect subject responses accurately. Looking at the log files, all RTs are 0 or 1 msec, indicating that the button response from the preceding trial is being taken as the response for the current trial.

So, contra my previous post, this is a HUGE problem!

thanks in advance for any help,
Aaron

[jarvis24]

Can you post a copy of your input file?

[AaronNewman]

Here it is.

Thanks for taking a look!
Aaron

[jarvis24]

Have you tried sending a signal at the end of each trial to set it to 0?

e.g., adding ttl:0, 1016, 0 to the end of each trial?