diff --git a/changelogs/fragments/57894-combine-filter-rework.yml b/changelogs/fragments/57894-combine-filter-rework.yml
new file mode 100644
index 00000000000..0652df34be5
--- /dev/null
+++ b/changelogs/fragments/57894-combine-filter-rework.yml
@@ -0,0 +1,4 @@
+bugfixes:
+- combine filter - ``[dict1, [dict2]] | combine`` now raise an error; previously ``combine`` had an undocumented behaviour where it was flattening the list before combining it (https://github.com/ansible/ansible/pull/57894#discussion_r339517518).
+minor_changes:
+- combine filter - now accept a ``list_merge`` argument which modifies its behaviour when the hashes to merge contain arrays/lists.
diff --git a/docs/docsite/rst/user_guide/playbooks_filters.rst b/docs/docsite/rst/user_guide/playbooks_filters.rst
index 3f734febb2f..3da2c29d944 100644
--- a/docs/docsite/rst/user_guide/playbooks_filters.rst
+++ b/docs/docsite/rst/user_guide/playbooks_filters.rst
@@ -345,12 +345,13 @@ You can use the transformed data with ``loop`` to iterate over the same subeleme
 
 .. _combine_filter:
 
-Combining hashes
-----------------
+Combining hashes/dictionaries
+-----------------------------
 
 .. versionadded:: 2.0
 
-The `combine` filter allows hashes to be merged. For example, the following would override keys in one hash::
+The ``combine`` filter allows hashes to be merged.
+For example, the following would override keys in one hash::
 
     {{ {'a':1, 'b':2} | combine({'b':3}) }}
 
@@ -358,25 +359,208 @@ The resulting hash would be::
 
     {'a':1, 'b':3}
 
-The filter also accepts an optional `recursive=True` parameter to not
-only override keys in the first hash, but also recurse into nested
-hashes and merge their keys too:
-
-.. code-block:: jinja
-
-    {{ {'a':{'foo':1, 'bar':2}, 'b':2} | combine({'a':{'bar':3, 'baz':4}}, recursive=True) }}
-
-This would result in::
-
-    {'a':{'foo':1, 'bar':3, 'baz':4}, 'b':2}
-
 The filter can also take multiple arguments to merge::
 
     {{ a | combine(b, c, d) }}
+    {{ [a, b, c, d] | combine }}
 
-In this case, keys in `d` would override those in `c`, which would override those in `b`, and so on.
+In this case, keys in ``d`` would override those in ``c``, which would
+override those in ``b``, and so on.
+
+The filter also accepts two optional parameters: ``recursive`` and ``list_merge``.
+
+recursive
+  Is a boolean, default to ``False``.
+  Should the ``combine`` recursively merge nested hashes.
+  Note: It does **not** depend on the value of the ``hash_behaviour`` setting in ``ansible.cfg``.
+
+list_merge
+  Is a string, its possible values are ``replace`` (default), ``keep``, ``append``, ``prepend``, ``append_rp`` or ``prepend_rp``.
+  It modifies the behaviour of ``combine`` when the hashes to merge contain arrays/lists.
+
+.. code-block:: yaml
+
+    default:
+      a:
+        x: default
+        y: default
+      b: default
+      c: default
+    patch:
+      a:
+        y: patch
+        z: patch
+      b: patch
+
+If ``recursive=False`` (the default), nested hash aren't merged::
+
+    {{ default | combine(patch) }}
+
+This would result in::
+
+    a:
+      y: patch
+      z: patch
+    b: patch
+    c: default
+
+If ``recursive=True``, recurse into nested hash and merge their keys::
+
+    {{ default | combine(patch, recursive=True) }}
+
+This would result in::
+
+    a:
+      x: default
+      y: patch
+      z: patch
+    b: patch
+    c: default
+
+If ``list_merge='replace'`` (the default), arrays from the right hash will "replace" the ones in the left hash::
+
+    default:
+      a:
+        - default
+    patch:
+      a:
+        - patch
+
+.. code-block:: jinja
+
+    {{ default | combine(patch) }}
+
+This would result in::
+
+    a:
+      - patch
+
+If ``list_merge='keep'``, arrays from the left hash will be kept::
+
+    {{ default | combine(patch, list_merge='keep') }}
+
+This would result in::
+
+    a:
+      - default
+
+If ``list_merge='append'``, arrays from the right hash will be appended to the ones in the left hash::
+
+    {{ default | combine(patch, list_merge='append') }}
+
+This would result in::
+
+    a:
+      - default
+      - patch
+
+If ``list_merge='prepend'``, arrays from the right hash will be prepended to the ones in the left hash::
+
+    {{ default | combine(patch, list_merge='prepend') }}
+
+This would result in::
+
+    a:
+      - patch
+      - default
+
+If ``list_merge='append_rp'``, arrays from the right hash will be appended to the ones in the left hash.
+Elements of arrays in the left hash that are also in the corresponding array of the right hash will be removed ("rp" stands for "remove present").
+Duplicate elements that aren't in both hashes are kept::
+
+    default:
+      a:
+        - 1
+        - 1
+        - 2
+        - 3
+    patch:
+      a:
+        - 3
+        - 4
+        - 5
+        - 5
+
+.. code-block:: jinja
+
+    {{ default | combine(patch, list_merge='append_rp') }}
+
+This would result in::
+
+    a:
+      - 1
+      - 1
+      - 2
+      - 3
+      - 4
+      - 5
+      - 5
+
+If ``list_merge='prepend_rp'``, the behavior is similar to the one for ``append_rp``, but elements of arrays in the right hash are prepended::
+
+    {{ default | combine(patch, list_merge='prepend_rp') }}
+
+This would result in::
+
+    a:
+      - 3
+      - 4
+      - 5
+      - 5
+      - 1
+      - 1
+      - 2
+
+``recursive`` and ``list_merge`` can be used together::
+
+    default:
+      a:
+        a':
+          x: default_value
+          y: default_value
+          list:
+            - default_value
+      b:
+        - 1
+        - 1
+        - 2
+        - 3
+    patch:
+      a:
+        a':
+          y: patch_value
+          z: patch_value
+          list:
+            - patch_value
+      b:
+        - 3
+        - 4
+        - 4
+        - key: value
+
+.. code-block:: jinja
+
+    {{ default | combine(patch, recursive=True, list_merge='append_rp') }}
+
+This would result in::
+
+    a:
+      a':
+        x: default_value
+        y: patch_value
+        z: patch_value
+        list:
+          - default_value
+          - patch_value
+    b:
+      - 1
+      - 1
+      - 2
+      - 3
+      - 4
+      - 4
+      - key: value
 
-This behavior does not depend on the value of the `hash_behavior` setting in `ansible.cfg`.
 
 .. _extract_filter:
 
diff --git a/lib/ansible/plugins/filter/core.py b/lib/ansible/plugins/filter/core.py
index aab48a009fd..d2ae68a2441 100644
--- a/lib/ansible/plugins/filter/core.py
+++ b/lib/ansible/plugins/filter/core.py
@@ -45,7 +45,7 @@ from ansible.module_utils.six import iteritems, string_types, integer_types, rer
 from ansible.module_utils.six.moves import reduce, shlex_quote
 from ansible.module_utils._text import to_bytes, to_native, to_text
 from ansible.module_utils.common.collections import is_sequence
-from ansible.module_utils.common._collections_compat import Mapping, MutableMapping
+from ansible.module_utils.common._collections_compat import Mapping
 from ansible.parsing.ajson import AnsibleJSONEncoder
 from ansible.parsing.yaml.dumper import AnsibleDumper
 from ansible.template import recursive_check_defined
@@ -306,25 +306,36 @@ def mandatory(a, msg=None):
 
 
 def combine(*terms, **kwargs):
-    recursive = kwargs.get('recursive', False)
-    if len(kwargs) > 1 or (len(kwargs) == 1 and 'recursive' not in kwargs):
-        raise AnsibleFilterError("'recursive' is the only valid keyword argument")
+    recursive = kwargs.pop('recursive', False)
+    list_merge = kwargs.pop('list_merge', 'replace')
+    if kwargs:
+        raise AnsibleFilterError("'recursive' and 'list_merge' are the only valid keyword arguments")
 
-    dicts = []
-    for t in terms:
-        if isinstance(t, MutableMapping):
-            recursive_check_defined(t)
-            dicts.append(t)
-        elif isinstance(t, list):
-            recursive_check_defined(t)
-            dicts.append(combine(*t, **kwargs))
-        else:
-            raise AnsibleFilterError("|combine expects dictionaries, got " + repr(t))
+    # allow the user to do `[dict1, dict2, ...] | combine`
+    dictionaries = flatten(terms, levels=1)
 
-    if recursive:
-        return reduce(merge_hash, dicts)
-    else:
-        return dict(itertools.chain(*map(iteritems, dicts)))
+    # recursively check that every elements are defined (for jinja2)
+    recursive_check_defined(dictionaries)
+
+    if not dictionaries:
+        return {}
+
+    if len(dictionaries) == 1:
+        return dictionaries[0]
+
+    # merge all the dicts so that the dict at the end of the array have precedence
+    # over the dict at the beginning.
+    # we merge the dicts from the highest to the lowest priority because there is
+    # a huge probability that the lowest priority dict will be the biggest in size
+    # (as the low prio dict will hold the "default" values and the others will be "patches")
+    # and merge_hash create a copy of it's first argument.
+    # so high/right -> low/left is more efficient than low/left -> high/right
+    high_to_low_prio_dict_iterator = reversed(dictionaries)
+    result = next(high_to_low_prio_dict_iterator)
+    for dictionary in high_to_low_prio_dict_iterator:
+        result = merge_hash(dictionary, result, recursive, list_merge)
+
+    return result
 
 
 def comment(text, style='plain', **kw):
diff --git a/lib/ansible/utils/vars.py b/lib/ansible/utils/vars.py
index b0d5bc5b798..9011b377798 100644
--- a/lib/ansible/utils/vars.py
+++ b/lib/ansible/utils/vars.py
@@ -31,7 +31,7 @@ from ansible import context
 from ansible.errors import AnsibleError, AnsibleOptionsError
 from ansible.module_utils.six import iteritems, string_types
 from ansible.module_utils._text import to_native, to_text
-from ansible.module_utils.common._collections_compat import MutableMapping
+from ansible.module_utils.common._collections_compat import MutableMapping, MutableSequence
 from ansible.parsing.splitter import parse_kv
 
 
@@ -92,32 +92,92 @@ def combine_vars(a, b):
         return result
 
 
-def merge_hash(a, b):
+def merge_hash(x, y, recursive=True, list_merge='replace'):
     """
-    Recursively merges hash b into a so that keys from b take precedence over keys from a
+    Return a new dictionary result of the merges of y into x,
+    so that keys from y take precedence over keys from x.
+    (x and y aren't modified)
     """
+    if list_merge not in ('replace', 'keep', 'append', 'prepend', 'append_rp', 'prepend_rp'):
+        raise AnsibleError("merge_hash: 'list_merge' argument can only be equal to 'replace', 'keep', 'append', 'prepend', 'append_rp' or 'prepend_rp'")
 
-    _validate_mutable_mappings(a, b)
+    # verify x & y are dicts
+    _validate_mutable_mappings(x, y)
 
-    # if a is empty or equal to b, return b
-    if a == {} or a == b:
-        return b.copy()
+    # to speed things up: if x is empty or equal to y, return y
+    # (this `if` can be remove without impact on the function
+    #  except performance)
+    if x == {} or x == y:
+        return y.copy()
 
-    # if b is empty the below unfolds quickly
-    result = a.copy()
+    # in the following we will copy elements from y to x, but
+    # we don't want to modify x, so we create a copy of it
+    x = x.copy()
 
-    # next, iterate over b keys and values
-    for k, v in iteritems(b):
-        # if there's already such key in a
-        # and that key contains a MutableMapping
-        if k in result and isinstance(result[k], MutableMapping) and isinstance(v, MutableMapping):
-            # merge those dicts recursively
-            result[k] = merge_hash(result[k], v)
-        else:
-            # otherwise, just copy the value from b to a
-            result[k] = v
+    # to speed things up: use dict.update if possible
+    # (this `if` can be remove without impact on the function
+    #  except performance)
+    if not recursive and list_merge == 'replace':
+        x.update(y)
+        return x
 
-    return result
+    # insert each element of y in x, overriding the one in x
+    # (as y has higher priority)
+    # we copy elements from y to x instead of x to y because
+    # there is a high probability x will be the "default" dict the user
+    # want to "patch" with y
+    # therefore x will have much more elements than y
+    for key, y_value in iteritems(y):
+        # if `key` isn't in x
+        # update x and move on to the next element of y
+        if key not in x:
+            x[key] = y_value
+            continue
+        # from this point we know `key` is in x
+
+        x_value = x[key]
+
+        # if both x's element and y's element are dicts
+        # recursively "combine" them or override x's with y's element
+        # depending on the `recursive` argument
+        # and move on to the next element of y
+        if isinstance(x_value, MutableMapping) and isinstance(y_value, MutableMapping):
+            if recursive:
+                x[key] = merge_hash(x_value, y_value, recursive, list_merge)
+            else:
+                x[key] = y_value
+            continue
+
+        # if both x's element and y's element are lists
+        # "merge" them depending on the `list_merge` argument
+        # and move on to the next element of y
+        if isinstance(x_value, MutableSequence) and isinstance(y_value, MutableSequence):
+            if list_merge == 'replace':
+                # replace x value by y's one as it has higher priority
+                x[key] = y_value
+            elif list_merge == 'append':
+                x[key] = x_value + y_value
+            elif list_merge == 'prepend':
+                x[key] = y_value + x_value
+            elif list_merge == 'append_rp':
+                # append all elements from y_value (high prio) to x_value (low prio)
+                # and remove x_value elements that are also in y_value
+                # we don't remove elements from x_value nor y_value that were already in double
+                # (we assume that there is a reason if there where such double elements)
+                # _rp stands for "remove present"
+                x[key] = [z for z in x_value if z not in y_value] + y_value
+            elif list_merge == 'prepend_rp':
+                # same as 'append_rp' but y_value elements are prepend
+                x[key] = y_value + [z for z in x_value if z not in y_value]
+            # else 'keep'
+            #   keep x value even if y it's of higher priority
+            #   it's done by not changing x[key]
+            continue
+
+        # else just override x's element with y's one
+        x[key] = y_value
+
+    return x
 
 
 def load_extra_vars(loader):
diff --git a/test/integration/targets/filter_core/tasks/main.yml b/test/integration/targets/filter_core/tasks/main.yml
index 2a1f8c76a4a..9db278dde80 100644
--- a/test/integration/targets/filter_core/tasks/main.yml
+++ b/test/integration/targets/filter_core/tasks/main.yml
@@ -136,6 +136,82 @@
     - "'Ansible - くらとみ\n' | b64encode(encoding='utf-16-le') == 'QQBuAHMAaQBiAGwAZQAgAC0AIABPMIkwaDB/MAoA'"
     - "'QQBuAHMAaQBiAGwAZQAgAC0AIABPMIkwaDB/MAoA' | b64decode(encoding='utf-16-le') == 'Ansible - くらとみ\n'"
 
+- set_fact:
+    x:
+      x: x
+      key: x
+    y:
+      y: y
+      key: y
+    z:
+      z: z
+      key: z
+
+    # Most complicated combine dicts from the documentation
+    default:
+      a:
+        a':
+          x: default_value
+          y: default_value
+          list:
+            - default_value
+      b:
+        - 1
+        - 1
+        - 2
+        - 3
+    patch:
+      a:
+        a':
+          y: patch_value
+          z: patch_value
+          list:
+            - patch_value
+      b:
+        - 3
+        - 4
+        - 4
+        - key: value
+    result:
+      a:
+        a':
+          x: default_value
+          y: patch_value
+          z: patch_value
+          list:
+            - default_value
+            - patch_value
+      b:
+        - 1
+        - 1
+        - 2
+        - 3
+        - 4
+        - 4
+        - key: value
+
+- name: Verify combine filter
+  assert:
+    that:
+      - "([x] | combine) == x"
+      - "(x | combine(y)) == {'x': 'x', 'y': 'y', 'key': 'y'}"
+      - "(x | combine(y, z)) == {'x': 'x', 'y': 'y', 'z': 'z', 'key': 'z'}"
+      - "([x, y, z] | combine) == {'x': 'x', 'y': 'y', 'z': 'z', 'key': 'z'}"
+      - "([x, y] | combine(z)) == {'x': 'x', 'y': 'y', 'z': 'z', 'key': 'z'}"
+      # more advance dicts combination tests are done in "merge_hash" function unit tests
+      # but even it's redundant with those unit tests, we do at least the most complicated exemple of the documentation here
+      - "(default | combine(patch, recursive=True, list_merge='append_rp')) == result"
+
+- set_fact:
+    combine: "{{[x, [y]] | combine(z)}}"
+  ignore_errors: yes
+  register: result
+
+- name: Ensure combining objects which aren't dictionaries throws an error
+  assert:
+    that:
+      - "result.msg.startswith(\"failed to combine variables, expected dicts but got\")"
+
 - name: Ensure combining two dictionaries containing undefined variables provides a helpful error
   block:
     - set_fact:
@@ -159,13 +235,3 @@
     - assert:
         that:
           - "result.msg.startswith('The task includes an option with an undefined variable')"
-
-    - set_fact:
-        key2: is_defined
-
-    - set_fact:
-        combined: "{{ foo | combine({'key2': key2}) }}"
-
-    - assert:
-        that:
-          - "combined.key2 == 'is_defined'"
diff --git a/test/units/utils/test_vars.py b/test/units/utils/test_vars.py
index 812c1670211..c92ce4b6ed8 100644
--- a/test/units/utils/test_vars.py
+++ b/test/units/utils/test_vars.py
@@ -28,12 +28,17 @@ from ansible.utils.vars import combine_vars, merge_hash
 
 
 class TestVariableUtils(unittest.TestCase):
+    def setUp(self):
+        pass
 
-    test_merge_data = (
+    def tearDown(self):
+        pass
+
+    combine_vars_merge_data = (
         dict(
             a=dict(a=1),
             b=dict(b=2),
-            result=dict(a=1, b=2)
+            result=dict(a=1, b=2),
         ),
         dict(
             a=dict(a=1, c=dict(foo='bar')),
@@ -46,7 +51,7 @@ class TestVariableUtils(unittest.TestCase):
             result=defaultdict(a=1, b=2, c=defaultdict(foo='bar', baz='bam'))
         ),
     )
-    test_replace_data = (
+    combine_vars_replace_data = (
         dict(
             a=dict(a=1),
             b=dict(b=2),
@@ -64,11 +69,7 @@ class TestVariableUtils(unittest.TestCase):
         ),
     )
 
-    def test_merge_hash(self):
-        for test in self.test_merge_data:
-            self.assertEqual(merge_hash(test['a'], test['b']), test['result'])
-
-    def test_improper_args(self):
+    def test_combine_vars_improper_args(self):
         with mock.patch('ansible.constants.DEFAULT_HASH_BEHAVIOUR', 'replace'):
             with self.assertRaises(AnsibleError):
                 combine_vars([1, 2, 3], dict(a=1))
@@ -83,10 +84,199 @@ class TestVariableUtils(unittest.TestCase):
 
     def test_combine_vars_replace(self):
         with mock.patch('ansible.constants.DEFAULT_HASH_BEHAVIOUR', 'replace'):
-            for test in self.test_replace_data:
+            for test in self.combine_vars_replace_data:
                 self.assertEqual(combine_vars(test['a'], test['b']), test['result'])
 
     def test_combine_vars_merge(self):
         with mock.patch('ansible.constants.DEFAULT_HASH_BEHAVIOUR', 'merge'):
-            for test in self.test_merge_data:
+            for test in self.combine_vars_merge_data:
                 self.assertEqual(combine_vars(test['a'], test['b']), test['result'])
+
+    merge_hash_data = {
+        "low_prio": {
+            "a": {
+                "a'": {
+                    "x": "low_value",
+                    "y": "low_value",
+                    "list": ["low_value"]
+                }
+            },
+            "b": [1, 1, 2, 3]
+        },
+        "high_prio": {
+            "a": {
+                "a'": {
+                    "y": "high_value",
+                    "z": "high_value",
+                    "list": ["high_value"]
+                }
+            },
+            "b": [3, 4, 4, {"5": "value"}]
+        }
+    }
+
+    def test_merge_hash_simple(self):
+        for test in self.combine_vars_merge_data:
+            self.assertEqual(merge_hash(test['a'], test['b']), test['result'])
+
+        low = self.merge_hash_data['low_prio']
+        high = self.merge_hash_data['high_prio']
+        expected = {
+            "a": {
+                "a'": {
+                    "x": "low_value",
+                    "y": "high_value",
+                    "z": "high_value",
+                    "list": ["high_value"]
+                }
+            },
+            "b": high['b']
+        }
+        self.assertEqual(merge_hash(low, high), expected)
+
+    def test_merge_hash_non_recursive_and_list_replace(self):
+        low = self.merge_hash_data['low_prio']
+        high = self.merge_hash_data['high_prio']
+        expected = high
+        self.assertEqual(merge_hash(low, high, False, 'replace'), expected)
+
+    def test_merge_hash_non_recursive_and_list_keep(self):
+        low = self.merge_hash_data['low_prio']
+        high = self.merge_hash_data['high_prio']
+        expected = {
+            "a": high['a'],
+            "b": low['b']
+        }
+        self.assertEqual(merge_hash(low, high, False, 'keep'), expected)
+
+    def test_merge_hash_non_recursive_and_list_append(self):
+        low = self.merge_hash_data['low_prio']
+        high = self.merge_hash_data['high_prio']
+        expected = {
+            "a": high['a'],
+            "b": low['b'] + high['b']
+        }
+        self.assertEqual(merge_hash(low, high, False, 'append'), expected)
+
+    def test_merge_hash_non_recursive_and_list_prepend(self):
+        low = self.merge_hash_data['low_prio']
+        high = self.merge_hash_data['high_prio']
+        expected = {
+            "a": high['a'],
+            "b": high['b'] + low['b']
+        }
+        self.assertEqual(merge_hash(low, high, False, 'prepend'), expected)
+
+    def test_merge_hash_non_recursive_and_list_append_rp(self):
+        low = self.merge_hash_data['low_prio']
+        high = self.merge_hash_data['high_prio']
+        expected = {
+            "a": high['a'],
+            "b": [1, 1, 2] + high['b']
+        }
+        self.assertEqual(merge_hash(low, high, False, 'append_rp'), expected)
+
+    def test_merge_hash_non_recursive_and_list_prepend_rp(self):
+        low = self.merge_hash_data['low_prio']
+        high = self.merge_hash_data['high_prio']
+        expected = {
+            "a": high['a'],
+            "b": high['b'] + [1, 1, 2]
+        }
+        self.assertEqual(merge_hash(low, high, False, 'prepend_rp'), expected)
+
+    def test_merge_hash_recursive_and_list_replace(self):
+        low = self.merge_hash_data['low_prio']
+        high = self.merge_hash_data['high_prio']
+        expected = {
+            "a": {
+                "a'": {
+                    "x": "low_value",
+                    "y": "high_value",
+                    "z": "high_value",
+                    "list": ["high_value"]
+                }
+            },
+            "b": high['b']
+        }
+        self.assertEqual(merge_hash(low, high, True, 'replace'), expected)
+
+    def test_merge_hash_recursive_and_list_keep(self):
+        low = self.merge_hash_data['low_prio']
+        high = self.merge_hash_data['high_prio']
+        expected = {
+            "a": {
+                "a'": {
+                    "x": "low_value",
+                    "y": "high_value",
+                    "z": "high_value",
+                    "list": ["low_value"]
+                }
+            },
+            "b": low['b']
+        }
+        self.assertEqual(merge_hash(low, high, True, 'keep'), expected)
+
+    def test_merge_hash_recursive_and_list_append(self):
+        low = self.merge_hash_data['low_prio']
+        high = self.merge_hash_data['high_prio']
+        expected = {
+            "a": {
+                "a'": {
+                    "x": "low_value",
+                    "y": "high_value",
+                    "z": "high_value",
+                    "list": ["low_value", "high_value"]
+                }
+            },
+            "b": low['b'] + high['b']
+        }
+        self.assertEqual(merge_hash(low, high, True, 'append'), expected)
+
+    def test_merge_hash_recursive_and_list_prepend(self):
+        low = self.merge_hash_data['low_prio']
+        high = self.merge_hash_data['high_prio']
+        expected = {
+            "a": {
+                "a'": {
+                    "x": "low_value",
+                    "y": "high_value",
+                    "z": "high_value",
+                    "list": ["high_value", "low_value"]
+                }
+            },
+            "b": high['b'] + low['b']
+        }
+        self.assertEqual(merge_hash(low, high, True, 'prepend'), expected)
+
+    def test_merge_hash_recursive_and_list_append_rp(self):
+        low = self.merge_hash_data['low_prio']
+        high = self.merge_hash_data['high_prio']
+        expected = {
+            "a": {
+                "a'": {
+                    "x": "low_value",
+                    "y": "high_value",
+                    "z": "high_value",
+                    "list": ["low_value", "high_value"]
+                }
+            },
+            "b": [1, 1, 2] + high['b']
+        }
+        self.assertEqual(merge_hash(low, high, True, 'append_rp'), expected)
+
+    def test_merge_hash_recursive_and_list_prepend_rp(self):
+        low = self.merge_hash_data['low_prio']
+        high = self.merge_hash_data['high_prio']
+        expected = {
+            "a": {
+                "a'": {
+                    "x": "low_value",
+                    "y": "high_value",
+                    "z": "high_value",
+                    "list": ["high_value", "low_value"]
+                }
+            },
+            "b": high['b'] + [1, 1, 2]
+        }
+        self.assertEqual(merge_hash(low, high, True, 'prepend_rp'), expected)