diff --git a/doc/wisdom/README.md b/doc/wisdom/README.md
new file mode 100644
index 0000000..6013b1c
--- /dev/null
+++ b/doc/wisdom/README.md
@@ -0,0 +1,58 @@
+# The Wisdom Improvement Protocol
+
+## How Self-Differentiation Unites AIs, Humans, and Societies in Better Decision-Making
+
+![diagram](./wisdom.png)
+
+## Overview
+
+The **Wisdom Improvement Protocol** offers a novel framework for understanding how AIs, humans, and societies can enhance decision-making through an integrated process of perception, evaluation, action, and reflection. By drawing on Karl Friston’s **active inference model**, this protocol emphasizes **self-differentiation**—the ability to stay true to core values while adapting to changing environments. This approach not only minimizes prediction errors but also aligns actions with deeper principles, fostering continuous growth and wisdom.
+
+## How It Works: A Step-by-Step Walkthrough
+
+### 1. Initial Context
+
+- **Context[n]:** Every decision-making process begins within a specific environment that presents stimuli, serving as the starting point for action. This context, in active inference terms, represents the sensory data that the system or individual must interpret to reduce uncertainty.
+
+### 2. Perception and Interpretation
+
+- **Schema (Sensor):** The stimuli are organized and categorized through a mental framework, or schema, which structures the information to make it actionable. This parallels the predictive coding process in active inference, where the system anticipates sensory inputs based on prior knowledge.
+- **Ontology (Classifier):** The categorized information is then interpreted to form concepts, refining the system’s understanding of the environment. This step ensures that new data aligns with existing models of the world.
+
+### 3. Emotional and Value-Based Processing
+
+- **Values (RightBrain):** The concepts are evaluated through an emotional or ethical lens, guided by core values and beliefs. This step adds depth to the decision-making process, ensuring that actions are not just about minimizing prediction errors but also about aligning with what matters most.
+
+### 4. Goal Formation and Decision-Making
+
+- **Goals (LeftBrain):** Based on the evaluation of values, specific goals are set, directing actions in a way that is consistent with both immediate needs and long-term principles. In active inference, this is akin to selecting actions that fulfill predictions while aligning with overarching objectives.
+
+### 5. Action and Response
+
+- **Capabilities (Body):** The decisions are translated into concrete actions, which impact the environment and generate feedback. These actions embody the system’s efforts to resolve prediction errors by actively engaging with the world, a core principle of active inference.
+
+### 6. Reflection and Adaptation
+
+- **Awareness (Mind):** After actions are taken, outcomes are reflected upon, leading to the reassessment of schemas, ontologies, values, and goals. This reflection phase is critical for adapting and updating internal models based on new experiences, ensuring that the system continues to evolve. The red lines in the diagram emphasize that this is the stage where modifications occur , driving continuous learning and adaptation -- though more easily at the higher levels.
+
+### 7. New Context
+
+- **Context[n+1]:** The environment is updated as a result of the actions, providing new stimuli and restarting the cycle. This ongoing process of action and reflection fosters the iterative learning essential to developing wisdom.
+
+## How These Features Implement Self-Differentiation
+
+### Embodied Wisdom
+
+The protocol shows that wisdom is more than just an intellectual exercise—it integrates mind, emotion, and action. Self-differentiation ensures that decisions—whether made by AIs, individuals, or societies—are grounded in core principles while effectively engaging with the external world. Through actions, wisdom becomes embodied in tangible outcomes, linking theory with practice.
+
+### Iterative Learning
+
+Wisdom evolves through a continuous cycle of experience, reflection, and adaptation. In this process, self-differentiation provides stability by keeping core elements like schemas, ontologies, values, and goals consistent during inference while allowing for their modification during reflection. This balance between consistency and adaptability is key to ongoing learning and improvement.
+
+### Contextual Response
+
+Effective decision-making requires an understanding of and response to specific contexts. In the active inference framework, this means minimizing prediction errors in a way that is adaptive to the environment. Self-differentiation ensures that decisions remain aligned with core values even as they adapt to new challenges, making them both principled and responsive.
+
+## Conclusion
+
+The **Wisdom Improvement Protocol** integrates Karl Friston’s active inference model with the concept of self-differentiation, offering a structured approach to decision-making that balances consistency with adaptability. By following this protocol, AIs, humans, and societies can make decisions that are not only effective in minimizing prediction errors but also deeply aligned with core values, fostering continuous growth and wisdom across diverse domains. This model provides a powerful framework for enhancing decision-making in technology, personal development, and societal governance.
diff --git a/doc/wisdom/wisdom.DOT b/doc/wisdom/wisdom.DOT
new file mode 100644
index 0000000..712056b
--- /dev/null
+++ b/doc/wisdom/wisdom.DOT
@@ -0,0 +1,72 @@
+digraph Wisdom {
+
+    label="Wisdom Improvement Protocol";
+    labelloc="t";
+
+    // Context nodes
+    Context_n [shape="diamond"; label="Context[n]";];
+    Context_n1 [shape="diamond"; label="Context[n+1]";];
+
+    subgraph cluster_Mind {
+        label = "Mind";
+        style = "filled";
+        color = "lightpink";
+        awareness [shape="box"];
+    }
+
+    // Define clusters for the parameters
+    subgraph cluster_Sensor {
+        label = "Sensor";
+        style = "filled";
+        color = "lightgrey";
+        schema [shape="box"];
+    }
+
+    subgraph cluster_Classifier {
+        label = "Classifier";
+        style = "filled";
+        color = "lightgrey";
+        ontology [shape="box"];
+    }
+
+    subgraph cluster_RightBrain {
+        label = "RightBrain";
+        style = "filled";
+        color = "lightgrey";
+        values [shape="box"];
+    }
+
+    subgraph cluster_LeftBrain {
+        label = "LeftBrain";
+        style = "filled";
+        color = "lightgrey";
+        goals [shape="box"];
+    }
+
+
+    // Place Mind and Body clusters on the same level with different colors
+    subgraph cluster_Body {
+        label = "Body";
+        style = "filled";
+        color = "lightgreen";  // Different color for Body cluster
+        capabilities [shape="box"];
+    }
+
+
+    // Main workflow connections
+    Context_n -> schema [label="stimulus"];
+    schema -> ontology [label="perception"];
+    ontology -> values [label="concept"];
+    values -> goals [label="emotion"];
+    goals -> capabilities [label="decision", color="green"];
+    capabilities -> Context_n1 [label="action", color="green", constraint=false];
+    Context_n1 -> schema [label="consequences", color="green", constraint=false];
+
+    // Reflection edges
+    goals -> awareness [label="reflection", color="blue", constraint=false];
+    awareness -> schema [color="red", style="dotted"];
+    awareness -> ontology [color="red", style="dashed"];
+    awareness -> values [color="red", style="solid"];
+    awareness -> goals [color="red", style="bold"];
+
+}
diff --git a/doc/wisdom/wisdom.png b/doc/wisdom/wisdom.png
new file mode 100644
index 0000000..85ff373
Binary files /dev/null and b/doc/wisdom/wisdom.png differ