diff --git a/skpro/model_selection/_tuning.py b/skpro/model_selection/_tuning.py
index 84f50dc07..d2a48b8cc 100644
--- a/skpro/model_selection/_tuning.py
+++ b/skpro/model_selection/_tuning.py
@@ -8,6 +8,7 @@
 import pandas as pd
 from sklearn.model_selection import ParameterGrid, ParameterSampler, check_cv
 
+from skpro.metrics import CRPS
 from skpro.benchmarking.evaluate import evaluate
 from skpro.regression.base._delegate import _DelegatedProbaRegressor
 from skpro.utils.parallel import parallelize
@@ -156,6 +157,8 @@ def _fit(self, X, y, C=None):
 
         # scoring = check_scoring(self.scoring, obj=self)
         scoring = self.scoring
+        if scoring is None:
+            scoring = CRPS()
         scoring_name = f"test_{scoring.name}"
 
         backend = self.backend
@@ -631,6 +634,8 @@ def get_test_params(cls, parameter_set="default"):
             "scoring": PinballLoss(),
             "error_score": "raise",
         }
+        param_no_backend = {**param1, "backend": "None"}
+        params = [param1, param2, params3, param_no_backend]
 
         from skpro.regression.online._refit import OnlineRefit
 
@@ -832,6 +837,7 @@ def __init__(
         verbose=0,
         return_n_best_estimators=1,
         random_state=None,
+        backend="loky",
         error_score=np.nan,
         backend_params=None,
         update_behaviour="no_update",
@@ -843,6 +849,7 @@ def __init__(
             cv=cv,
             verbose=verbose,
             return_n_best_estimators=return_n_best_estimators,
+            backend=backend,
             error_score=error_score,
             backend_params=backend_params,
             update_behaviour=update_behaviour,
@@ -906,6 +913,8 @@ def get_test_params(cls, parameter_set="default"):
             "scoring": PinballLoss(),
             "error_score": "raise",
         }
+        param_no_backend = {**param1, "backend": "None"}
+        params = [param1, param2, params3, param_no_backend]
 
         from skpro.regression.online._refit import OnlineRefit
 
diff --git a/skpro/model_selection/tests/__init__.py b/skpro/model_selection/tests/__init__.py
new file mode 100644
index 000000000..90ba1208e
--- /dev/null
+++ b/skpro/model_selection/tests/__init__.py
@@ -0,0 +1 @@
+"""Tests for model selection utilities."""
\ No newline at end of file
diff --git a/skpro/model_selection/tests/test_tuning.py b/skpro/model_selection/tests/test_tuning.py
new file mode 100644
index 000000000..7adaaee85
--- /dev/null
+++ b/skpro/model_selection/tests/test_tuning.py
@@ -0,0 +1,37 @@
+"""Tests for model selection tuning utilities."""
+
+import pandas as pd
+import pytest
+from sklearn.model_selection import KFold
+
+from skpro.model_selection import GridSearchCV, RandomizedSearchCV
+from skpro.regression.dummy import DummyProbaRegressor
+from skpro.tests.test_switch import run_test_module_changed
+
+
+def _get_test_data():
+    X = pd.DataFrame({"x": [1, 2, 3, 4]})
+    y = pd.DataFrame({"y": [1.0, 2.0, 3.0, 4.0]})
+    return X, y
+
+
+@pytest.mark.skipif(
+    not run_test_module_changed("skpro.model_selection"),
+    reason="Test only if skpro.model_selection has been changed",
+)
+def test_randomizedsearch_accepts_backend_argument():
+    """RandomizedSearchCV should accept backend argument documented in API."""
+    X, y = _get_test_data()
+
+    rscv = RandomizedSearchCV(
+        estimator=DummyProbaRegressor(),
+        cv=KFold(n_splits=2),
+        param_distributions={"strategy": ["empirical"]},
+        n_iter=1,
+        scoring=None,
+        backend="threading",
+        backend_params={"n_jobs": 1},
+    )
+    rscv.fit(X, y)
+
+    assert rscv.backend == "threading"
\ No newline at end of file