search

int-brain-lab / iblrig

Main repository for IBL rig code
MIT License
21 stars 20 forks source link

Estimate impact of reverting the stimOn extraction to first pulse after stimOnTrigger #654

Closed k1o0 closed 1 week ago

k1o0 commented 2 months ago

Pre v5 stim on times were extracted as the first f2ttl change after the trigger times (here). Now the stim on time is simply the first f2ttl signal of the trial, however badly calibrated devices and a likely bug in the Bonsai which causes the sync square to change when it shouldn't(!) means that the stim on time is often wrong.

We should assess the impact of reverting to the old way (e.g. n sessions, n trials affected). This should improve session QC outcomes.

k1o0 commented 1 month ago

0/489 would get worse. 489/489 would stay the same. 110/489 would improve. 0 overall outcomes would improve.

k1o0 commented 1 week ago

See https://docs.google.com/presentation/d/1r2LsKF-79nkkqC4kzVKx3qIDYANdqWfMRtBlr-GlPZY/edit#slide=id.g2e1ace90223_1_0