==============================================================================
POWER SIMULATION -- naive_boxa_d6_13   [metric: # successes (count)]
==============================================================================
model: behavior ~ stim + day + stim:day + (1|rat)   (behavior = # successes (count); day within-window)
observed groups: stim n=3, control n=7   nrep=120, alpha=0.05
ground truth: stim:day=+0.8737/day, rat SD=10.75, residual SD=11.53, days=8

 true stim:day interaction = +0.8737 (100% of observed)
  N/group  | per-animal power | LME power
  ------------------------------------------
  3        |  0.07            |  0.11  <- observed
  5        |  0.17            |  0.21
  8        |  0.17            |  0.19
  12       |  0.23            |  0.23
  16       |  0.28            |  0.33
  24       |  0.38            |  0.41

 true stim:day interaction = +0.4368 (50% of observed)
  N/group  | per-animal power | LME power
  ------------------------------------------
  3        |  0.06            |  0.10  <- observed
  5        |  0.05            |  0.08
  8        |  0.07            |  0.11
  12       |  0.06            |  0.08
  16       |  0.07            |  0.07
  24       |  0.16            |  0.15

Read the per-animal column as the honest power; LME is optimistic (obs-level DF).
