diff --git a/DESCRIPTION b/DESCRIPTION
index a2bf90f45..f32873fbb 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -1,7 +1,7 @@
 Type: Package
 Package: parameters
 Title: Processing of Model Parameters
-Version: 0.22.2.9
+Version: 0.22.2.10
 Authors@R:
     c(person(given = "Daniel",
              family = "Lüdecke",
diff --git a/NAMESPACE b/NAMESPACE
index f7bbef008..ae1aca375 100644
--- a/NAMESPACE
+++ b/NAMESPACE
@@ -378,6 +378,7 @@ S3method(p_significance,lm)
 S3method(p_significance,lme)
 S3method(p_significance,merMod)
 S3method(p_significance,mixed)
+S3method(p_significance,parameters_model)
 S3method(p_significance,rma)
 S3method(p_significance,svyglm)
 S3method(p_significance,wbm)
diff --git a/R/p_significance.R b/R/p_significance.R
index d22e06e7d..603e7ed5a 100644
--- a/R/p_significance.R
+++ b/R/p_significance.R
@@ -182,8 +182,27 @@ p_significance.lm <- function(x, threshold = "default", ci = 0.95, verbose = TRU
 
 .posterior_ci <- function(x, ci, ...) {
   # first, we need CIs
-  out <- ci(x, ci = ci, ...)
-  dof <- .safe(insight::get_df(x, type = "wald"), Inf)
+  if (inherits(x, "parameters_model")) {
+    # for model_parameters objects, directly extract CIs
+    out <- as.data.frame(x)[intersect(
+      c("Parameter", "CI_low", "CI_high", "Component", "Effects"),
+      colnames(x)
+    )]
+    ci <- attributes(x)$ci
+    # and extract degrees of freedom
+    df_column <- grep("(df|df_error)", colnames(x))
+    if (length(df_column) > 0) {
+      dof <- unique(x[[df_column]])
+      if (length(dof) > 1) {
+        dof <- Inf
+      }
+    } else {
+      dof <- Inf
+    }
+  } else {
+    out <- ci(x, ci = ci, ...)
+    dof <- .safe(insight::get_df(x, type = "wald"), Inf)
+  }
   # we now iterate all confidence intervals and create an approximate normal
   # distribution that covers the CI-range.
   posterior <- as.data.frame(lapply(seq_len(nrow(out)), function(i) {
@@ -281,3 +300,6 @@ p_significance.zeroinfl <- p_significance.lm
 
 #' @export
 p_significance.rma <- p_significance.lm
+
+#' @export
+p_significance.parameters_model <- p_significance.lm
diff --git a/tests/testthat/test-p_significance.R b/tests/testthat/test-p_significance.R
index 092abf8bf..9de21d3cc 100644
--- a/tests/testthat/test-p_significance.R
+++ b/tests/testthat/test-p_significance.R
@@ -10,6 +10,11 @@ test_that("p_significance", {
   expect_named(x, c("Parameter", "CI", "CI_low", "CI_high", "ps"))
   expect_snapshot(print(x))
 
+  mp <- model_parameters(m)
+  set.seed(123)
+  x2 <- p_significance(mp)
+  expect_equal(x$ps, x2$ps, tolerance = 1e-4)
+
   set.seed(123)
   x <- p_significance(m, ci = 0.8)
   expect_equal(x$ps, c(1, 0.3983, 0.9959, 0.6188, 0), tolerance = 1e-3)