Next, to ensure that non-significant MI values did not inflate the estimates of SE, the p values were used to calculate a weight (w) for each MI result via w = -log10(p). These weights were then normalized by dividing each weight by the sum of all the weights and used to calculate the weighted mean (Eq. 3) and SE of the weighted mean (Eq. 4) across all relevant neurons (animal strain and liquid type) at a given time bin t.(3)I¯t(x,y)=∑iwiIi,t(x,y). (4)SEMt(x,y)=σt∑iwi2.