Picture-Word-N1/01 Validation/02 Analysis/02_fit_bayesian.R

library(brms)

dat_rt <- readRDS("dat_rt.rds")

n_cores <- 7

seed <- 42
n_iter <- 25000
n_warmup <- 17500
adapt_delta <- 0.99
max_treedepth <- 11
n_chains <- 5

priors <- c(
  # FIXED EFFECTS
  set_prior("normal(0,7.5)", class = "b", coef = "Intercept"),
  set_prior("normal(0,2.5)", class = "b", coef = "cong_dev"),
  set_prior("normal(0,2.5)", class = "b", coef = "pred_norm"),
  set_prior("normal(0,2.5)", class = "b", coef = "cong_dev:pred_norm"),
  set_prior("normal(0,7.5)", class = "b", coef = "Intercept", dpar="sigma"),
  set_prior("normal(0,2.5)", class = "b", coef = "cong_dev", dpar="sigma"),
  set_prior("normal(0,2.5)", class = "b", coef = "pred_norm", dpar="sigma"),
  set_prior("normal(0,2.5)", class = "b", coef = "cong_dev:pred_norm", dpar="sigma"),
  set_prior("normal(0,7.5)", class = "b", coef = "Intercept", dpar="ndt"),
  set_prior("normal(0,2.5)", class = "b", coef = "cong_dev", dpar="ndt"),
  set_prior("normal(0,2.5)", class = "b", coef = "pred_norm", dpar="ndt"),
  set_prior("normal(0,2.5)", class = "b", coef = "cong_dev:pred_norm", dpar="ndt"),
  # STANDARD DEVIATIONS OF RANDOM EFFECT DISTRIBUTIONS
  set_prior("student_t(3, 0, 2)", class = "sd"),
  set_prior("student_t(3, 0, 2)", class = "sd", dpar="sigma"),
  set_prior("student_t(3, 0, 2)", class = "sd", dpar="ndt")
)

# Fit model

m_bme <- brm(
  brmsformula(
    rt ~ 0 + Intercept + cong_dev * pred_norm +
      (cong_dev * pred_norm | subj_id) +
      (cong_dev | image) +
      (1 | string),
    sigma ~ 0 + Intercept + cong_dev * pred_norm +
      (cong_dev * pred_norm | subj_id) +
      (cong_dev | image) +
      (1 | string),
    ndt ~ 0 + Intercept + cong_dev * pred_norm +
      (cong_dev * pred_norm | subj_id) +
      (cong_dev | image) +
      (1 | string)
  ),
  data = dat_rt,
  family = shifted_lognormal(),
  prior = priors,
  inits = "0",
  iter = n_iter,
  warmup = n_warmup,
  chains = n_chains,
  control = list(
    adapt_delta = adapt_delta,
    max_treedepth = max_treedepth
  ),
  sample_prior = "no",
  silent = TRUE,
  cores = n_cores,
  seed = seed,
  thin = 1,
  file = file.path("mods", "m_bme.rds")
)