Difference in summary tables between `brms` and `mgcv` GAMM

Interfaces brms
1

Following on this post HERE, can anyone explain the difference in summary tables from mgcv and brms please?

For example:

gam1 <-gam(DV ~
    ExpTrialOrder +
    FactorA +
    s(CovariateM) +
    s(ExpTrialOrder, Participant, bs='fs', m=1),
    ...)

brms_gam1 <-brm(DV ~
    ExpTrialOrder +
    FactorA +
    s(CovariateM) +
    s(ExpTrialOrder, Participant, bs='fs', m=1),
    ...)

would give something like this:

# mgcv
# Approximate significance of smooth terms:
#                               edf Ref.df     F p-value
# ...
# s(TrialOrder.z,Participant) 93.58    399 2.218  <2e-16

# brms
# Smooth Terms: 
#                               Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
# ...
# sds(szTrialOrderSubject_nr_1)     0.31      0.05     0.22     0.41 1.00     3389     5414
# sds(szTrialOrderSubject_nr_2)     2.07      0.26     1.62     2.63 1.00     3734     5053

How can one compare the factor smooth outputs from these two tables? And what does each line from brms, with _1 and _2 mean?

  • Operating System: macOS Ventura
  • brms Version: 2.19.0