Hey! Were you able to resolve that problem? I seem to have a similar problem at the moment and posted on the forum a few days ago. This is the link to my post. For me, in the beginning everything seems to work out fine, but the problem just gets progressively worse with each experimental block. I would really appreciate any kind of idea where the problem is stemming from as we already scheduled scanning slots and i need “correct” logging of responses to calculate blockwise performance-dependent rewards.