diff --git a/pandas/core/arrays/datetimelike.py b/pandas/core/arrays/datetimelike.py
index ba5bfc108f16b..bb40cf78ea006 100644
--- a/pandas/core/arrays/datetimelike.py
+++ b/pandas/core/arrays/datetimelike.py
@@ -183,7 +183,7 @@ def _rebox_native(cls, value: int) -> Union[int, np.datetime64, np.timedelta64]:
         """
         raise AbstractMethodError(cls)
 
-    def _unbox_scalar(self, value: DTScalarOrNaT) -> int:
+    def _unbox_scalar(self, value: DTScalarOrNaT, setitem: bool = False) -> int:
         """
         Unbox the integer value of a scalar `value`.
 
@@ -191,6 +191,8 @@ def _unbox_scalar(self, value: DTScalarOrNaT) -> int:
         ----------
         value : Period, Timestamp, Timedelta, or NaT
             Depending on subclass.
+        setitem : bool, default False
+            Whether to check compatiblity with setitem strictness.
 
         Returns
         -------
@@ -841,6 +843,7 @@ def _validate_listlike(
             if is_dtype_equal(value.categories.dtype, self.dtype):
                 # TODO: do we need equal dtype or just comparable?
                 value = value._internal_get_values()
+                value = extract_array(value, extract_numpy=True)
 
         if allow_object and is_object_dtype(value.dtype):
             pass
@@ -875,8 +878,7 @@ def _validate_setitem_value(self, value):
             # TODO: cast_str for consistency?
             value = self._validate_scalar(value, msg, cast_str=False)
 
-        self._check_compatible_with(value, setitem=True)
-        return self._unbox(value)
+        return self._unbox(value, setitem=True)
 
     def _validate_insert_value(self, value):
         msg = f"cannot insert {type(self).__name__} with incompatible label"
@@ -886,6 +888,8 @@ def _validate_insert_value(self, value):
         # TODO: if we dont have compat, should we raise or astype(object)?
         #  PeriodIndex does astype(object)
         return value
+        # Note: we do not unbox here because the caller needs boxed value
+        #  to check for freq.
 
     def _validate_where_value(self, other):
         msg = f"Where requires matching dtype, not {type(other)}"
@@ -893,20 +897,18 @@ def _validate_where_value(self, other):
             other = self._validate_scalar(other, msg)
         else:
             other = self._validate_listlike(other, "where")
-            self._check_compatible_with(other, setitem=True)
 
-        self._check_compatible_with(other, setitem=True)
-        return self._unbox(other)
+        return self._unbox(other, setitem=True)
 
-    def _unbox(self, other) -> Union[np.int64, np.ndarray]:
+    def _unbox(self, other, setitem: bool = False) -> Union[np.int64, np.ndarray]:
         """
         Unbox either a scalar with _unbox_scalar or an instance of our own type.
         """
         if lib.is_scalar(other):
-            other = self._unbox_scalar(other)
+            other = self._unbox_scalar(other, setitem=setitem)
         else:
             # same type as self
-            self._check_compatible_with(other)
+            self._check_compatible_with(other, setitem=setitem)
             other = other.view("i8")
         return other
 
diff --git a/pandas/core/arrays/datetimes.py b/pandas/core/arrays/datetimes.py
index 9f10cc84dcfcc..56e0a9861548f 100644
--- a/pandas/core/arrays/datetimes.py
+++ b/pandas/core/arrays/datetimes.py
@@ -451,11 +451,11 @@ def _generate_range(
     def _rebox_native(cls, value: int) -> np.datetime64:
         return np.int64(value).view("M8[ns]")
 
-    def _unbox_scalar(self, value):
+    def _unbox_scalar(self, value, setitem: bool = False):
         if not isinstance(value, self._scalar_type) and value is not NaT:
             raise ValueError("'value' should be a Timestamp.")
         if not isna(value):
-            self._check_compatible_with(value)
+            self._check_compatible_with(value, setitem=setitem)
         return value.value
 
     def _scalar_from_string(self, value):
diff --git a/pandas/core/arrays/period.py b/pandas/core/arrays/period.py
index eea11bde77030..865b1680c008a 100644
--- a/pandas/core/arrays/period.py
+++ b/pandas/core/arrays/period.py
@@ -257,11 +257,13 @@ def _generate_range(cls, start, end, periods, freq, fields):
     def _rebox_native(cls, value: int) -> np.int64:
         return np.int64(value)
 
-    def _unbox_scalar(self, value: Union[Period, NaTType]) -> int:
+    def _unbox_scalar(
+        self, value: Union[Period, NaTType], setitem: bool = False
+    ) -> int:
         if value is NaT:
             return value.value
         elif isinstance(value, self._scalar_type):
-            self._check_compatible_with(value)
+            self._check_compatible_with(value, setitem=setitem)
             return value.ordinal
         else:
             raise ValueError(f"'value' should be a Period. Got '{value}' instead.")
diff --git a/pandas/core/arrays/timedeltas.py b/pandas/core/arrays/timedeltas.py
index 5e3c0f2b8d876..3eaf428bc64b2 100644
--- a/pandas/core/arrays/timedeltas.py
+++ b/pandas/core/arrays/timedeltas.py
@@ -283,10 +283,10 @@ def _generate_range(cls, start, end, periods, freq, closed=None):
     def _rebox_native(cls, value: int) -> np.timedelta64:
         return np.int64(value).view("m8[ns]")
 
-    def _unbox_scalar(self, value):
+    def _unbox_scalar(self, value, setitem: bool = False):
         if not isinstance(value, self._scalar_type) and value is not NaT:
             raise ValueError("'value' should be a Timedelta.")
-        self._check_compatible_with(value)
+        self._check_compatible_with(value, setitem=setitem)
         return value.value
 
     def _scalar_from_string(self, value):
diff --git a/pandas/tests/arrays/test_datetimelike.py b/pandas/tests/arrays/test_datetimelike.py
index d2d3766959fbf..9d316c38082af 100644
--- a/pandas/tests/arrays/test_datetimelike.py
+++ b/pandas/tests/arrays/test_datetimelike.py
@@ -244,7 +244,7 @@ def test_searchsorted(self):
         # GH#29884 match numpy convention on whether NaT goes
         #  at the end or the beginning
         result = arr.searchsorted(pd.NaT)
-        if _np_version_under1p18 or self.array_cls is PeriodArray:
+        if np_version_under1p18 or self.array_cls is PeriodArray:
             # Following numpy convention, NaT goes at the beginning
             #  (unlike NaN which goes at the end)
             assert result == 0