==============================================================================
LOG-DAY MODEL + COHEN'S f + POWER -- matched_current_d0_3   [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=3, control n=3   nrep=120, alpha=0.05

--- fitted on real data ---
stim x log(day) interaction: F(1,20)=7.974  p(resid)=0.01049  p(Satt)=0.01125 (df=18)
  honest per-animal random slope (log-day): F(1,18.3)=7.99  p=0.01106
Cohen's f (interaction, partial eta^2=0.157) = 0.431  (large; f: .10 small, .25 medium, .40 large)

--- power simulation (log-day ground truth: stim:logday=+0.1785, ratSD=0.01747, resSD=0.08062) ---

 true stim:log(day) = +0.1785 (100% of observed)
  N/group  | per-animal power | LME power
  ------------------------------------------
  3        |  0.42            |  0.82  <- observed
  5        |  0.88            |  0.95
  8        |  1.00            |  0.99
  12       |  1.00            |  1.00
  16       |  1.00            |  1.00
  24       |  1.00            |  1.00

 true stim:log(day) = +0.08926 (50% of observed)
  N/group  | per-animal power | LME power
  ------------------------------------------
  3        |  0.12            |  0.26  <- observed
  5        |  0.39            |  0.50
  8        |  0.53            |  0.63
  12       |  0.82            |  0.82
  16       |  0.87            |  0.90
  24       |  0.97            |  0.98

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