Cool thanks. Lots of ideas but variance is easiest to look at for a precision/accuracy tradeoff. The use of a variable start time may also have affected it.
That's right, we checked it visually, which showed larger variance in the action condition in variable-foreperiod experiments, but we did not run any tests. Because this difference between initiation conditions did not appear to interact with foreperiod duration, we chose not to focus on it.
Comments
I am pretty sure @alexandrenobre.bsky.social looked at it, but I don't think we did formal testing on it. Do you have an idea of what's going on?
I was positive it was something about action improving learning, but, in general, sequential effects did not change much with/without action.