diff --git a/docs/misc/changelog.rst b/docs/misc/changelog.rst
index a53fd999..113ce3f9 100644
--- a/docs/misc/changelog.rst
+++ b/docs/misc/changelog.rst
@@ -2,6 +2,7 @@
 
 Changelog
 ==========
+- Renamed environment output observations in ``evaluate_policy`` to prevent shadowing the input observations during callbacks (@npit)
 
 Release 1.8.0 (2023-04-07)
 --------------------------
@@ -418,4 +419,4 @@ Contributors:
 -------------
 
 @ku2482 @guyk1971 @minhlong94 @ayeright @kronion @glmcdona @cyprienc @sgillen @Gregwar @rnederstigt @qgallouedec
-@mlodel @CppMaster @burakdmb @honglu2875 @ZikangXiong @AlexPasqua @jonasreiher
+@mlodel @CppMaster @burakdmb @honglu2875 @ZikangXiong @AlexPasqua @jonasreiher @npit
diff --git a/sb3_contrib/common/maskable/evaluation.py b/sb3_contrib/common/maskable/evaluation.py
index 8a9fba83..0330c28b 100644
--- a/sb3_contrib/common/maskable/evaluation.py
+++ b/sb3_contrib/common/maskable/evaluation.py
@@ -103,7 +103,7 @@ def evaluate_policy(
             actions, states = model.predict(
                 observations, state=states, episode_start=episode_starts, deterministic=deterministic
             )
-        observations, rewards, dones, infos = env.step(actions)
+        new_observations, rewards, dones, infos = env.step(actions)
         current_rewards += rewards
         current_lengths += 1
         for i in range(n_envs):
@@ -116,6 +116,7 @@ def evaluate_policy(
 
                 if callback is not None:
                     callback(locals(), globals())
+                observations = new_observations
 
                 if dones[i]:
                     if is_monitor_wrapped: