==============================================================================
LOG-DAY MODEL + COHEN'S f + POWER -- right_only_d6_13   [metric: success RATE]
==============================================================================
model: behavior ~ stim + log(day) + stim:log(day) + (1|rat)   (behavior = success RATE)
log(day) uses 1-indexed training day (our day 0 = paper "Day 1")
observed groups: stim n=4, control n=2   nrep=120, alpha=0.05

--- fitted on real data ---
stim x log(day) interaction: F(1,38)=4.975  p(resid)=0.0317  p(Satt)=0.03178 (df=38)
  honest per-animal random slope (log-day): F(1,4.9)=0.25  p=0.6396
Cohen's f (interaction, partial eta^2=0.039) = 0.201  (small-medium; f: .10 small, .25 medium, .40 large)

--- power simulation (log-day ground truth: stim:logday=+0.1853, ratSD=0.04282, resSD=0.05196) ---

 true stim:log(day) = +0.1853 (100% of observed)
  N/group  | per-animal power | LME power
  ------------------------------------------
  3        |  0.53            |  0.80
  5        |  0.88            |  0.96
  8        |  0.99            |  1.00
  12       |  1.00            |  1.00
  16       |  1.00            |  1.00
  24       |  1.00            |  1.00

 true stim:log(day) = +0.09266 (50% of observed)
  N/group  | per-animal power | LME power
  ------------------------------------------
  3        |  0.16            |  0.27
  5        |  0.31            |  0.42
  8        |  0.52            |  0.62
  12       |  0.74            |  0.79
  16       |  0.86            |  0.89
  24       |  0.98            |  0.99

Read per-animal as the honest power; LME matches the paper's power code (optimistic).
