diff --git a/README.md b/README.md
index d23ffdadeb9701630cf4fd29fbb50de860e1d5f3..e07212f9a004fbcf8768f52f7161876948547288 100644
--- a/README.md
+++ b/README.md
@@ -251,7 +251,6 @@ tools and refine the plan based on feedback from the environment.
     *Theodore R. Sumers, Shunyu Yao, Karthik Narasimhan, Thomas L. Griffiths (2024)*
     Published on *arXiv*
 
-
 Cognitive abilities once considered unique to humans—such as reasoning,
 planning, and reflection, along with a degree of self-control, self-awareness,
 and self-improvement—can now be achieved by leveraging well-crafted prompts in
@@ -262,6 +261,14 @@ LLMs combined with embedded cognitive intelligence.
   Ilhan, Selim Tekin, Gaowen Liu, Ramana Kompella, Ling Liu (2024) Published on
   arXiv
 
+An LLM can operate within a game as a player, a non-player character, a game
+master, a designer or an analyst.
+
+- **[Large Language Models and Games: A Survey and
+  Roadmap](http://dx.doi.org/10.1109/TG.2024.3461510)** Roberto Gallotta, Graham
+  Todd, Marvin Zammit, Sam Earle, Antonios Liapis, Julian Togelius, Georgios N.
+  Yannakakis (2024) Published in *IEEE Transactions on Games*
+
 LLMs have the ability to emulate a real human in certain experiments in
 experimental economics or social psychology.