diff --git a/hindsight-api/alembic/versions/e0a1b2c3d4e5_disposition_to_3_traits.py b/hindsight-api/alembic/versions/e0a1b2c3d4e5_disposition_to_3_traits.py
new file mode 100644
index 00000000..6712f4cc
--- /dev/null
+++ b/hindsight-api/alembic/versions/e0a1b2c3d4e5_disposition_to_3_traits.py
@@ -0,0 +1,62 @@
+"""disposition_to_3_traits
+
+Revision ID: e0a1b2c3d4e5
+Revises: rename_personality
+Create Date: 2024-12-08
+
+Migrate disposition traits from Big Five (openness, conscientiousness, extraversion,
+agreeableness, neuroticism, bias_strength with 0-1 float values) to the new 3-trait
+system (skepticism, literalism, empathy with 1-5 integer values).
+"""
+from typing import Sequence, Union
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision: str = 'e0a1b2c3d4e5'
+down_revision: Union[str, Sequence[str], None] = 'rename_personality'
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    """Convert Big Five disposition to 3-trait disposition."""
+    conn = op.get_bind()
+
+    # Update all existing banks to use the new disposition format
+    # Convert from old format to new format with reasonable mappings:
+    # - skepticism: derived from inverse of agreeableness (skeptical people are less agreeable)
+    # - literalism: derived from conscientiousness (detail-oriented people are more literal)
+    # - empathy: derived from agreeableness + inverse of neuroticism
+    # Default all to 3 (neutral) for simplicity
+    conn.execute(sa.text("""
+        UPDATE banks
+        SET disposition = '{"skepticism": 3, "literalism": 3, "empathy": 3}'::jsonb
+        WHERE disposition IS NOT NULL
+    """))
+
+    # Update the default for new banks
+    conn.execute(sa.text("""
+        ALTER TABLE banks
+        ALTER COLUMN disposition SET DEFAULT '{"skepticism": 3, "literalism": 3, "empathy": 3}'::jsonb
+    """))
+
+
+def downgrade() -> None:
+    """Convert back to Big Five disposition."""
+    conn = op.get_bind()
+
+    # Revert to Big Five format with default values
+    conn.execute(sa.text("""
+        UPDATE banks
+        SET disposition = '{"openness": 0.5, "conscientiousness": 0.5, "extraversion": 0.5, "agreeableness": 0.5, "neuroticism": 0.5, "bias_strength": 0.5}'::jsonb
+        WHERE disposition IS NOT NULL
+    """))
+
+    # Update the default for new banks
+    conn.execute(sa.text("""
+        ALTER TABLE banks
+        ALTER COLUMN disposition SET DEFAULT '{"openness": 0.5, "conscientiousness": 0.5, "extraversion": 0.5, "agreeableness": 0.5, "neuroticism": 0.5, "bias_strength": 0.5}'::jsonb
+    """))
diff --git a/hindsight-api/hindsight_api/api/http.py b/hindsight-api/hindsight_api/api/http.py
index 92634137..44d6f126 100644
--- a/hindsight-api/hindsight_api/api/http.py
+++ b/hindsight-api/hindsight_api/api/http.py
@@ -439,24 +439,18 @@ class BanksResponse(BaseModel):
 
 
 class DispositionTraits(BaseModel):
-    """Disposition traits based on Big Five model."""
+    """Disposition traits that influence how memories are formed and interpreted."""
     model_config = ConfigDict(json_schema_extra={
         "example": {
-            "openness": 0.8,
-            "conscientiousness": 0.6,
-            "extraversion": 0.5,
-            "agreeableness": 0.7,
-            "neuroticism": 0.3,
-            "bias_strength": 0.7
+            "skepticism": 3,
+            "literalism": 3,
+            "empathy": 3
         }
     })
 
-    openness: float = Field(ge=0.0, le=1.0, description="Openness to experience (0-1)")
-    conscientiousness: float = Field(ge=0.0, le=1.0, description="Conscientiousness (0-1)")
-    extraversion: float = Field(ge=0.0, le=1.0, description="Extraversion (0-1)")
-    agreeableness: float = Field(ge=0.0, le=1.0, description="Agreeableness (0-1)")
-    neuroticism: float = Field(ge=0.0, le=1.0, description="Neuroticism (0-1)")
-    bias_strength: float = Field(ge=0.0, le=1.0, description="How strongly disposition influences opinions (0-1)")
+    skepticism: int = Field(ge=1, le=5, description="How skeptical vs trusting (1=trusting, 5=skeptical)")
+    literalism: int = Field(ge=1, le=5, description="How literally to interpret information (1=flexible, 5=literal)")
+    empathy: int = Field(ge=1, le=5, description="How much to consider emotional context (1=detached, 5=empathetic)")
 
 
 class BankProfileResponse(BaseModel):
@@ -466,12 +460,9 @@ class BankProfileResponse(BaseModel):
             "bank_id": "user123",
             "name": "Alice",
             "disposition": {
-                "openness": 0.8,
-                "conscientiousness": 0.6,
-                "extraversion": 0.5,
-                "agreeableness": 0.7,
-                "neuroticism": 0.3,
-                "bias_strength": 0.7
+                "skepticism": 3,
+                "literalism": 3,
+                "empathy": 3
             },
             "background": "I am a software engineer with 10 years of experience in startups"
         }
@@ -500,7 +491,7 @@ class AddBackgroundRequest(BaseModel):
     content: str = Field(description="New background information to add or merge")
     update_disposition: bool = Field(
         default=True,
-        description="If true, infer Big Five disposition traits from the merged background (default: true)"
+        description="If true, infer disposition traits from the merged background (default: true)"
     )
 
 
@@ -510,12 +501,9 @@ class BackgroundResponse(BaseModel):
         "example": {
             "background": "I was born in Texas. I am a software engineer with 10 years of experience.",
             "disposition": {
-                "openness": 0.7,
-                "conscientiousness": 0.6,
-                "extraversion": 0.5,
-                "agreeableness": 0.8,
-                "neuroticism": 0.4,
-                "bias_strength": 0.6
+                "skepticism": 3,
+                "literalism": 3,
+                "empathy": 3
             }
         }
     })
@@ -543,12 +531,9 @@ class BankListResponse(BaseModel):
                     "bank_id": "user123",
                     "name": "Alice",
                     "disposition": {
-                        "openness": 0.5,
-                        "conscientiousness": 0.5,
-                        "extraversion": 0.5,
-                        "agreeableness": 0.5,
-                        "neuroticism": 0.5,
-                        "bias_strength": 0.5
+                        "skepticism": 3,
+                        "literalism": 3,
+                        "empathy": 3
                     },
                     "background": "I am a software engineer",
                     "created_at": "2024-01-15T10:30:00Z",
@@ -567,12 +552,9 @@ class CreateBankRequest(BaseModel):
         "example": {
             "name": "Alice",
             "disposition": {
-                "openness": 0.8,
-                "conscientiousness": 0.6,
-                "extraversion": 0.5,
-                "agreeableness": 0.7,
-                "neuroticism": 0.3,
-                "bias_strength": 0.7
+                "skepticism": 3,
+                "literalism": 3,
+                "empathy": 3
             },
             "background": "I am a creative software engineer with 10 years of experience"
         }
@@ -1605,7 +1587,7 @@ This operation cannot be undone.
         "/v1/default/banks/{bank_id}/profile",
         response_model=BankProfileResponse,
         summary="Update memory bank disposition",
-        description="Update bank's Big Five disposition traits and bias strength",
+        description="Update bank's disposition traits (skepticism, literalism, empathy)",
         operation_id="update_bank_disposition"
     )
     async def api_update_bank_disposition(bank_id: str,
@@ -1852,7 +1834,7 @@ This operation cannot be undone.
         "/v1/default/banks/{bank_id}/memories",
         response_model=DeleteResponse,
         summary="Clear memory bank memories",
-        description="Delete memory units for a memory bank. Optionally filter by type (world, experience, opinion) to delete only specific types. This is a destructive operation that cannot be undone. The bank profile (personality and background) will be preserved.",
+        description="Delete memory units for a memory bank. Optionally filter by type (world, experience, opinion) to delete only specific types. This is a destructive operation that cannot be undone. The bank profile (disposition and background) will be preserved.",
         operation_id="clear_bank_memories"
     )
     async def api_clear_bank_memories(bank_id: str,
diff --git a/hindsight-api/hindsight_api/engine/memory_engine.py b/hindsight-api/hindsight_api/engine/memory_engine.py
index b5521faa..8b3d7414 100644
--- a/hindsight-api/hindsight_api/engine/memory_engine.py
+++ b/hindsight-api/hindsight_api/engine/memory_engine.py
@@ -959,7 +959,7 @@ class MemoryEngine:
         budget_mapping = {
             Budget.LOW: 100,
             Budget.MID: 300,
-            Budget.HIGH: 600
+            Budget.HIGH: 1000
         }
         thinking_budget = budget_mapping[budget]
 
@@ -2502,14 +2502,14 @@ Guidelines:
     async def update_bank_disposition(
         self,
         bank_id: str,
-        disposition: Dict[str, float]
+        disposition: Dict[str, int]
     ) -> None:
         """
         Update bank disposition traits.
 
         Args:
             bank_id: bank IDentifier
-            disposition: Dict with Big Five traits + bias_strength (all 0-1)
+            disposition: Dict with skepticism, literalism, empathy (all 1-5)
         """
         pool = await self._get_pool()
         await bank_utils.update_bank_disposition(pool, bank_id, disposition)
@@ -2997,22 +2997,23 @@ Guidelines:
                     )
 
                     if not entity_exists:
-                        logger.debug(f"[OBSERVATIONS] Entity {entity_id} not yet in bank {bank_id}, skipping")
                         continue
 
                     entity_name = entity_exists['canonical_name']
 
-                    # Count facts linked to this entity
+                    # Count facts linked to this entity (in this bank)
                     fact_count = await conn.fetchval(
-                        "SELECT COUNT(*) FROM unit_entities WHERE entity_id = $1",
-                        entity_uuid
+                        """
+                        SELECT COUNT(*) FROM unit_entities ue
+                        JOIN memory_units mu ON ue.unit_id = mu.id
+                        WHERE ue.entity_id = $1 AND mu.bank_id = $2
+                        """,
+                        entity_uuid, bank_id
                     ) or 0
 
                     # Only regenerate if entity has enough facts
                     if fact_count >= min_facts:
                         await self.regenerate_entity_observations(bank_id, entity_id, entity_name, version=None)
-                    else:
-                        logger.debug(f"[OBSERVATIONS] Skipping {entity_name} ({fact_count} facts < {min_facts} threshold)")
 
                 except Exception as e:
                     logger.error(f"[OBSERVATIONS] Error processing entity {entity_id}: {e}")
diff --git a/hindsight-api/hindsight_api/engine/response_models.py b/hindsight-api/hindsight_api/engine/response_models.py
index b6ce70b0..1cf09c48 100644
--- a/hindsight-api/hindsight_api/engine/response_models.py
+++ b/hindsight-api/hindsight_api/engine/response_models.py
@@ -12,25 +12,22 @@ from pydantic import BaseModel, Field, ConfigDict
 
 class DispositionTraits(BaseModel):
     """
-    Disposition traits for a bank using the Big Five model.
+    Disposition traits for a memory bank.
 
-    All traits are scored 0.0-1.0 where higher values indicate stronger presence of the trait.
+    All traits are scored 1-5 where:
+    - skepticism: 1=trusting, 5=skeptical (how much to doubt or question information)
+    - literalism: 1=flexible interpretation, 5=literal interpretation (how strictly to interpret information)
+    - empathy: 1=detached, 5=empathetic (how much to consider emotional context)
     """
-    openness: float = Field(description="Openness to experience (0.0-1.0)")
-    conscientiousness: float = Field(description="Conscientiousness and organization (0.0-1.0)")
-    extraversion: float = Field(description="Extraversion and sociability (0.0-1.0)")
-    agreeableness: float = Field(description="Agreeableness and cooperation (0.0-1.0)")
-    neuroticism: float = Field(description="Emotional sensitivity and neuroticism (0.0-1.0)")
-    bias_strength: float = Field(description="How strongly disposition influences thinking (0.0-1.0)")
+    skepticism: int = Field(ge=1, le=5, description="How skeptical vs trusting (1=trusting, 5=skeptical)")
+    literalism: int = Field(ge=1, le=5, description="How literally to interpret information (1=flexible, 5=literal)")
+    empathy: int = Field(ge=1, le=5, description="How much to consider emotional context (1=detached, 5=empathetic)")
 
     model_config = ConfigDict(json_schema_extra={
         "example": {
-            "openness": 0.8,
-            "conscientiousness": 0.6,
-            "extraversion": 0.4,
-            "agreeableness": 0.7,
-            "neuroticism": 0.3,
-            "bias_strength": 0.5
+            "skepticism": 3,
+            "literalism": 3,
+            "empathy": 3
         }
     })
 
diff --git a/hindsight-api/hindsight_api/engine/retain/bank_utils.py b/hindsight-api/hindsight_api/engine/retain/bank_utils.py
index 01d3b6de..8f6be868 100644
--- a/hindsight-api/hindsight_api/engine/retain/bank_utils.py
+++ b/hindsight-api/hindsight_api/engine/retain/bank_utils.py
@@ -13,12 +13,9 @@ from ..response_models import DispositionTraits
 logger = logging.getLogger(__name__)
 
 DEFAULT_DISPOSITION = {
-    "openness": 0.5,
-    "conscientiousness": 0.5,
-    "extraversion": 0.5,
-    "agreeableness": 0.5,
-    "neuroticism": 0.5,
-    "bias_strength": 0.5,
+    "skepticism": 3,
+    "literalism": 3,
+    "empathy": 3,
 }
 
 
@@ -32,7 +29,7 @@ class BankProfile(TypedDict):
 class BackgroundMergeResponse(BaseModel):
     """LLM response for background merge with disposition inference."""
     background: str = Field(description="Merged background in first person perspective")
-    disposition: DispositionTraits = Field(description="Inferred Big Five disposition traits")
+    disposition: DispositionTraits = Field(description="Inferred disposition traits (skepticism, literalism, empathy)")
 
 
 async def get_bank_profile(pool, bank_id: str) -> BankProfile:
@@ -92,7 +89,7 @@ async def get_bank_profile(pool, bank_id: str) -> BankProfile:
 async def update_bank_disposition(
     pool,
     bank_id: str,
-    disposition: Dict[str, float]
+    disposition: Dict[str, int]
 ) -> None:
     """
     Update bank disposition traits.
@@ -100,7 +97,7 @@ async def update_bank_disposition(
     Args:
         pool: Database connection pool
         bank_id: bank IDentifier
-        disposition: Dict with Big Five traits + bias_strength (all 0-1)
+        disposition: Dict with skepticism, literalism, empathy (all 1-5)
     """
     # Ensure bank exists first
     await get_bank_profile(pool, bank_id)
@@ -223,13 +220,10 @@ Instructions:
 3. Keep additions that don't conflict
 4. Output in FIRST PERSON ("I") perspective
 5. Be concise - keep merged background under 500 characters
-6. Infer Big Five disposition traits from the merged background:
-   - Openness: 0.0-1.0 (creativity, curiosity, openness to new ideas)
-   - Conscientiousness: 0.0-1.0 (organization, discipline, goal-directed)
-   - Extraversion: 0.0-1.0 (sociability, assertiveness, energy from others)
-   - Agreeableness: 0.0-1.0 (cooperation, empathy, consideration)
-   - Neuroticism: 0.0-1.0 (emotional sensitivity, anxiety, stress response)
-   - Bias Strength: 0.0-1.0 (how much disposition influences opinions)
+6. Infer disposition traits from the merged background (each 1-5 integer):
+   - Skepticism: 1-5 (1=trusting, takes things at face value; 5=skeptical, questions everything)
+   - Literalism: 1-5 (1=flexible interpretation, reads between lines; 5=literal, exact interpretation)
+   - Empathy: 1-5 (1=detached, focuses on facts; 5=empathetic, considers emotional context)
 
 CRITICAL: You MUST respond with ONLY a valid JSON object. No markdown, no code blocks, no explanations. Just the JSON.
 
@@ -237,22 +231,19 @@ Format:
 {{
   "background": "the merged background text in first person",
   "disposition": {{
-    "openness": 0.7,
-    "conscientiousness": 0.6,
-    "extraversion": 0.5,
-    "agreeableness": 0.8,
-    "neuroticism": 0.4,
-    "bias_strength": 0.6
+    "skepticism": 3,
+    "literalism": 3,
+    "empathy": 3
   }}
 }}
 
 Trait inference examples:
-- "creative artist" → openness: 0.8+, bias_strength: 0.6
-- "organized engineer" → conscientiousness: 0.8+, openness: 0.5-0.6
-- "startup founder" → openness: 0.8+, extraversion: 0.7+, neuroticism: 0.3-0.4
-- "risk-averse analyst" → openness: 0.3-0.4, conscientiousness: 0.8+, neuroticism: 0.6+
-- "rational and diligent" → conscientiousness: 0.7+, openness: 0.6+
-- "passionate and dramatic" → extraversion: 0.7+, neuroticism: 0.6+, openness: 0.7+"""
+- "I'm a lawyer" → skepticism: 4, literalism: 5, empathy: 2
+- "I'm a therapist" → skepticism: 2, literalism: 2, empathy: 5
+- "I'm an engineer" → skepticism: 3, literalism: 4, empathy: 3
+- "I've been burned before by trusting people" → skepticism: 5, literalism: 3, empathy: 3
+- "I try to understand what people really mean" → skepticism: 3, literalism: 2, empathy: 4
+- "I take contracts very seriously" → skepticism: 4, literalism: 5, empathy: 2"""
     else:
         prompt = f"""You are helping maintain a memory bank's background/profile.
 
@@ -349,13 +340,12 @@ Merged background:"""
 
             # Validate disposition values
             disposition = result.get("disposition", {})
-            for key in ["openness", "conscientiousness", "extraversion",
-                       "agreeableness", "neuroticism", "bias_strength"]:
+            for key in ["skepticism", "literalism", "empathy"]:
                 if key not in disposition:
-                    disposition[key] = 0.5  # Default to neutral
+                    disposition[key] = 3  # Default to neutral
                 else:
-                    # Clamp to [0, 1]
-                    disposition[key] = max(0.0, min(1.0, float(disposition[key])))
+                    # Clamp to [1, 5] and convert to int
+                    disposition[key] = max(1, min(5, int(disposition[key])))
 
             result["disposition"] = disposition
 
diff --git a/hindsight-api/hindsight_api/engine/retain/entity_processing.py b/hindsight-api/hindsight_api/engine/retain/entity_processing.py
index fea40657..d238951a 100644
--- a/hindsight-api/hindsight_api/engine/retain/entity_processing.py
+++ b/hindsight-api/hindsight_api/engine/retain/entity_processing.py
@@ -7,7 +7,7 @@ import logging
 from typing import List, Tuple, Dict, Any
 from uuid import UUID
 
-from .types import ProcessedFact, EntityRef
+from .types import ProcessedFact, EntityRef, EntityLink
 from . import link_utils
 
 logger = logging.getLogger(__name__)
@@ -20,7 +20,7 @@ async def process_entities_batch(
     unit_ids: List[str],
     facts: List[ProcessedFact],
     log_buffer: List[str] = None
-) -> List[Tuple[str, str, float]]:
+) -> List[EntityLink]:
     """
     Process entities for all facts and create entity links.
 
@@ -39,7 +39,7 @@ async def process_entities_batch(
         log_buffer: Optional buffer for detailed logging
 
     Returns:
-        List of entity link tuples: (unit_id, entity_id, confidence)
+        List of EntityLink objects for batch insertion
     """
     if not unit_ids or not facts:
         return []
@@ -75,14 +75,14 @@ async def process_entities_batch(
 
 async def insert_entity_links_batch(
     conn,
-    entity_links: List[Tuple[str, str, float]]
+    entity_links: List[EntityLink]
 ) -> None:
     """
     Insert entity links in batch.
 
     Args:
         conn: Database connection
-        entity_links: List of (unit_id, entity_id, confidence) tuples
+        entity_links: List of EntityLink objects
     """
     if not entity_links:
         return
diff --git a/hindsight-api/hindsight_api/engine/retain/fact_storage.py b/hindsight-api/hindsight_api/engine/retain/fact_storage.py
index 280de832..442098ff 100644
--- a/hindsight-api/hindsight_api/engine/retain/fact_storage.py
+++ b/hindsight-api/hindsight_api/engine/retain/fact_storage.py
@@ -118,7 +118,7 @@ async def ensure_bank_exists(conn, bank_id: str) -> None:
         SET updated_at = NOW()
         """,
         bank_id,
-        '{"openness": 0.5, "conscientiousness": 0.5, "extraversion": 0.5, "agreeableness": 0.5, "neuroticism": 0.5, "bias_strength": 0.5}',
+        '{"skepticism": 3, "literalism": 3, "empathy": 3}',
         ""
     )
 
diff --git a/hindsight-api/hindsight_api/engine/retain/link_utils.py b/hindsight-api/hindsight_api/engine/retain/link_utils.py
index 77f684bb..b358fd7f 100644
--- a/hindsight-api/hindsight_api/engine/retain/link_utils.py
+++ b/hindsight-api/hindsight_api/engine/retain/link_utils.py
@@ -6,6 +6,9 @@ import time
 import logging
 from typing import List
 from datetime import timedelta, datetime, timezone
+from uuid import UUID
+
+from .types import EntityLink
 
 logger = logging.getLogger(__name__)
 
@@ -305,10 +308,14 @@ async def extract_entities_batch_optimized(
         # Only link each new unit to the most recent MAX_LINKS_PER_ENTITY units
         MAX_LINKS_PER_ENTITY = 50  # Limit to prevent explosion when entity appears in many facts
         link_gen_start = time.time()
-        links = []
+        links: List[EntityLink] = []
         new_unit_set = set(unit_ids)  # Units from this batch
 
+        def to_uuid(val) -> UUID:
+            return UUID(val) if isinstance(val, str) else val
+
         for entity_id, units_with_entity in entity_to_units.items():
+            entity_uuid = to_uuid(entity_id)
             # Separate new units (from this batch) and existing units
             new_units = [u for u in units_with_entity if str(u) in new_unit_set or u in new_unit_set]
             existing_units = [u for u in units_with_entity if str(u) not in new_unit_set and u not in new_unit_set]
@@ -318,15 +325,15 @@ async def extract_entities_batch_optimized(
             new_units_to_link = new_units[-MAX_LINKS_PER_ENTITY:] if len(new_units) > MAX_LINKS_PER_ENTITY else new_units
             for i, unit_id_1 in enumerate(new_units_to_link):
                 for unit_id_2 in new_units_to_link[i+1:]:
-                    links.append((unit_id_1, unit_id_2, 'entity', 1.0, entity_id))
-                    links.append((unit_id_2, unit_id_1, 'entity', 1.0, entity_id))
+                    links.append(EntityLink(from_unit_id=to_uuid(unit_id_1), to_unit_id=to_uuid(unit_id_2), entity_id=entity_uuid))
+                    links.append(EntityLink(from_unit_id=to_uuid(unit_id_2), to_unit_id=to_uuid(unit_id_1), entity_id=entity_uuid))
 
             # Link new units to LIMITED existing units (most recent)
             existing_to_link = existing_units[-MAX_LINKS_PER_ENTITY:]  # Take most recent
             for new_unit in new_units:
                 for existing_unit in existing_to_link:
-                    links.append((new_unit, existing_unit, 'entity', 1.0, entity_id))
-                    links.append((existing_unit, new_unit, 'entity', 1.0, entity_id))
+                    links.append(EntityLink(from_unit_id=to_uuid(new_unit), to_unit_id=to_uuid(existing_unit), entity_id=entity_uuid))
+                    links.append(EntityLink(from_unit_id=to_uuid(existing_unit), to_unit_id=to_uuid(new_unit), entity_id=entity_uuid))
 
         _log(log_buffer, f"      [6.3.3] Generate {len(links)} links: {time.time() - link_gen_start:.3f}s", level='debug')
         _log(log_buffer, f"  [6.3] Entity link creation: {len(links)} links for {len(all_entity_ids)} unique entities in {time.time() - substep_start:.3f}s", level='debug')
@@ -546,7 +553,7 @@ async def create_semantic_links_batch(
         raise
 
 
-async def insert_entity_links_batch(conn, links: List[tuple], chunk_size: int = 50000):
+async def insert_entity_links_batch(conn, links: List[EntityLink], chunk_size: int = 50000):
     """
     Insert all entity links using COPY to temp table + INSERT for maximum speed.
 
@@ -556,7 +563,7 @@ async def insert_entity_links_batch(conn, links: List[tuple], chunk_size: int =
 
     Args:
         conn: Database connection
-        links: List of tuples (from_unit_id, to_unit_id, link_type, weight, entity_id)
+        links: List of EntityLink objects
         chunk_size: Number of rows per batch (default 50000)
     """
     if not links:
@@ -585,16 +592,16 @@ async def insert_entity_links_batch(conn, links: List[tuple], chunk_size: int =
     await conn.execute("TRUNCATE _temp_entity_links")
     logger.debug(f"      [9.2] Truncate temp table: {time_mod.time() - truncate_start:.3f}s")
 
-    # Convert links to proper format for COPY
+    # Convert EntityLink objects to tuples for COPY
     convert_start = time_mod.time()
     records = []
-    for from_id, to_id, link_type, weight, entity_id in links:
+    for link in links:
         records.append((
-            uuid_mod.UUID(from_id) if isinstance(from_id, str) else from_id,
-            uuid_mod.UUID(to_id) if isinstance(to_id, str) else to_id,
-            link_type,
-            weight,
-            uuid_mod.UUID(str(entity_id)) if entity_id and not isinstance(entity_id, uuid_mod.UUID) else entity_id
+            link.from_unit_id,
+            link.to_unit_id,
+            link.link_type,
+            link.weight,
+            link.entity_id
         ))
     logger.debug(f"      [9.3] Convert {len(records)} records: {time_mod.time() - convert_start:.3f}s")
 
diff --git a/hindsight-api/hindsight_api/engine/retain/orchestrator.py b/hindsight-api/hindsight_api/engine/retain/orchestrator.py
index 687f09dc..9687e12d 100644
--- a/hindsight-api/hindsight_api/engine/retain/orchestrator.py
+++ b/hindsight-api/hindsight_api/engine/retain/orchestrator.py
@@ -17,7 +17,7 @@ def utcnow():
     """Get current UTC time."""
     return datetime.now(timezone.utc)
 
-from .types import RetainContent, ExtractedFact, ProcessedFact
+from .types import RetainContent, ExtractedFact, ProcessedFact, EntityLink
 from . import (
     fact_extraction,
     embedding_processing,
@@ -373,7 +373,7 @@ async def _trigger_background_tasks(
     bank_id: str,
     unit_ids: List[str],
     facts: List[ProcessedFact],
-    entity_links: List,
+    entity_links: List[EntityLink],
     log_buffer: List[str] = None
 ) -> None:
     """Trigger opinion reinforcement and observation regeneration (sync)."""
@@ -388,19 +388,27 @@ async def _trigger_background_tasks(
             'unit_entities': fact_entities
         })
 
-    # Regenerate observations synchronously for top entities
+    # Regenerate observations synchronously for top entities by fact count
     TOP_N_ENTITIES = 5
     MIN_FACTS_THRESHOLD = 5
 
     if entity_links and regenerate_observations_fn:
-        unique_entity_ids = set()
+        # Count mentions per entity in this batch
+        entity_mention_counts: Dict[str, int] = {}
         for link in entity_links:
-            # links are tuples: (from_unit_id, to_unit_id, link_type, weight, entity_id)
-            if len(link) >= 5 and link[4]:
-                unique_entity_ids.add(str(link[4]))
+            if link.entity_id:
+                entity_id = str(link.entity_id)
+                entity_mention_counts[entity_id] = entity_mention_counts.get(entity_id, 0) + 1
+
+        if entity_mention_counts:
+            # Sort by mention count descending and take top N
+            sorted_entities = sorted(
+                entity_mention_counts.items(),
+                key=lambda x: x[1],
+                reverse=True
+            )
+            entities_to_process = [e[0] for e in sorted_entities[:TOP_N_ENTITIES]]
 
-        if unique_entity_ids:
-            entities_to_process = list(unique_entity_ids)[:TOP_N_ENTITIES]
             obs_start = time.time()
             # Run observation regeneration synchronously
             await regenerate_observations_fn(
diff --git a/hindsight-api/hindsight_api/engine/retain/types.py b/hindsight-api/hindsight_api/engine/retain/types.py
index 21b30a09..ff65efd4 100644
--- a/hindsight-api/hindsight_api/engine/retain/types.py
+++ b/hindsight-api/hindsight_api/engine/retain/types.py
@@ -176,6 +176,20 @@ class ProcessedFact:
         )
 
 
+@dataclass
+class EntityLink:
+    """
+    Link between two memory units through a shared entity.
+
+    Used for entity-based graph connections in the memory graph.
+    """
+    from_unit_id: UUID
+    to_unit_id: UUID
+    entity_id: UUID
+    link_type: str = 'entity'
+    weight: float = 1.0
+
+
 @dataclass
 class RetainBatch:
     """
diff --git a/hindsight-api/hindsight_api/engine/search/think_utils.py b/hindsight-api/hindsight_api/engine/search/think_utils.py
index 1a539fe2..cbbb14e9 100644
--- a/hindsight-api/hindsight_api/engine/search/think_utils.py
+++ b/hindsight-api/hindsight_api/engine/search/think_utils.py
@@ -28,30 +28,48 @@ class OpinionExtractionResponse(BaseModel):
     )
 
 
-def describe_trait(name: str, value: float) -> str:
-    """Convert trait value to descriptive text."""
-    if value >= 0.8:
-        return f"very high {name}"
-    elif value >= 0.6:
-        return f"high {name}"
-    elif value >= 0.4:
-        return f"moderate {name}"
-    elif value >= 0.2:
-        return f"low {name}"
-    else:
-        return f"very low {name}"
+def describe_trait_level(value: int) -> str:
+    """Convert trait value (1-5) to descriptive text."""
+    levels = {
+        1: "very low",
+        2: "low",
+        3: "moderate",
+        4: "high",
+        5: "very high"
+    }
+    return levels.get(value, "moderate")
 
 
 def build_disposition_description(disposition: DispositionTraits) -> str:
     """Build a disposition description string from disposition traits."""
-    return f"""Your disposition traits:
-- {describe_trait('openness to new ideas', disposition.openness)}
-- {describe_trait('conscientiousness and organization', disposition.conscientiousness)}
-- {describe_trait('extraversion and sociability', disposition.extraversion)}
-- {describe_trait('agreeableness and cooperation', disposition.agreeableness)}
-- {describe_trait('emotional sensitivity', disposition.neuroticism)}
+    skepticism_desc = {
+        1: "You are very trusting and tend to take information at face value.",
+        2: "You tend to trust information but may question obvious inconsistencies.",
+        3: "You have a balanced approach to information, neither too trusting nor too skeptical.",
+        4: "You are somewhat skeptical and often question the reliability of information.",
+        5: "You are highly skeptical and critically examine all information for accuracy and hidden motives."
+    }
 
-Disposition influence strength: {int(disposition.bias_strength * 100)}% (how much your disposition shapes your opinions)"""
+    literalism_desc = {
+        1: "You interpret information very flexibly, reading between the lines and inferring intent.",
+        2: "You tend to consider context and implied meaning alongside literal statements.",
+        3: "You balance literal interpretation with contextual understanding.",
+        4: "You prefer to interpret information more literally and precisely.",
+        5: "You interpret information very literally and focus on exact wording and commitments."
+    }
+
+    empathy_desc = {
+        1: "You focus primarily on facts and data, setting aside emotional context.",
+        2: "You consider facts first but acknowledge emotional factors exist.",
+        3: "You balance factual analysis with emotional understanding.",
+        4: "You give significant weight to emotional context and human factors.",
+        5: "You strongly consider the emotional state and circumstances of others when forming memories."
+    }
+
+    return f"""Your disposition traits:
+- Skepticism ({describe_trait_level(disposition.skepticism)}): {skepticism_desc.get(disposition.skepticism, skepticism_desc[3])}
+- Literalism ({describe_trait_level(disposition.literalism)}): {literalism_desc.get(disposition.literalism, literalism_desc[3])}
+- Empathy ({describe_trait_level(disposition.empathy)}): {empathy_desc.get(disposition.empathy, empathy_desc[3])}"""
 
 
 def format_facts_for_prompt(facts: List[MemoryFact]) -> str:
@@ -141,13 +159,28 @@ Based on everything I know, believe, and who I am (including my name, dispositio
 
 def get_system_message(disposition: DispositionTraits) -> str:
     """Get the system message for the think LLM call."""
-    bias_strength = disposition.bias_strength
-    if bias_strength >= 0.7:
-        disposition_instruction = "Your disposition strongly influences your thinking. Let your traits guide how you interpret facts and form opinions. Don't be afraid to be biased based on your disposition."
-    elif bias_strength >= 0.4:
-        disposition_instruction = "Your disposition moderately influences your thinking. Balance your personal traits with objective analysis."
-    else:
-        disposition_instruction = "Your disposition has minimal influence on your thinking. Focus primarily on facts while keeping your traits in mind."
+    # Build disposition-specific instructions based on trait values
+    instructions = []
+
+    # Skepticism influences how much to question/doubt information
+    if disposition.skepticism >= 4:
+        instructions.append("Be skeptical of claims and look for potential issues or inconsistencies.")
+    elif disposition.skepticism <= 2:
+        instructions.append("Trust the information provided and take statements at face value.")
+
+    # Literalism influences interpretation style
+    if disposition.literalism >= 4:
+        instructions.append("Interpret information literally and focus on exact commitments and wording.")
+    elif disposition.literalism <= 2:
+        instructions.append("Read between the lines and consider implied meaning and context.")
+
+    # Empathy influences consideration of emotional factors
+    if disposition.empathy >= 4:
+        instructions.append("Consider the emotional state and circumstances behind the information.")
+    elif disposition.empathy <= 2:
+        instructions.append("Focus on facts and outcomes rather than emotional context.")
+
+    disposition_instruction = " ".join(instructions) if instructions else "Balance your disposition traits when interpreting information."
 
     return f"You are a person with your own thoughts, experiences, opinions, and disposition. When asked a question, respond naturally like a human would - share what you genuinely think based on what you know and have experienced. {disposition_instruction} Be direct, express your views confidently, and use 'I think', 'I believe', 'in my view', etc. Respond in plain text without markdown formatting."
 
diff --git a/hindsight-api/hindsight_api/models.py b/hindsight-api/hindsight_api/models.py
index b587910b..fe481306 100644
--- a/hindsight-api/hindsight_api/models.py
+++ b/hindsight-api/hindsight_api/models.py
@@ -292,8 +292,7 @@ class Bank(Base):
         JSONB,
         nullable=False,
         server_default=sql_text(
-            '\'{"openness": 0.5, "conscientiousness": 0.5, "extraversion": 0.5, '
-            '"agreeableness": 0.5, "neuroticism": 0.5, "bias_strength": 0.5}\'::jsonb'
+            '\'{"skepticism": 3, "literalism": 3, "empathy": 3}\'::jsonb'
         )
     )
     background: Mapped[str] = mapped_column(Text, nullable=False, server_default="")
diff --git a/hindsight-api/tests/test_agents_api.py b/hindsight-api/tests/test_agents_api.py
index 8d960cde..6bab3dd3 100644
--- a/hindsight-api/tests/test_agents_api.py
+++ b/hindsight-api/tests/test_agents_api.py
@@ -1,5 +1,5 @@
 """
-Tests for agent management API (profile, personality, background).
+Tests for agent management API (profile, disposition, background).
 """
 import pytest
 import uuid
@@ -18,51 +18,42 @@ class TestAgentProfile:
 
     @pytest.mark.asyncio
     async def test_get_agent_profile_creates_default(self, memory: MemoryEngine):
-        """Test that getting a profile for a new agent creates default personality."""
+        """Test that getting a profile for a new agent creates default disposition."""
         bank_id = unique_agent_id("test_profile_default")
 
         profile = await memory.get_bank_profile(bank_id)
 
         assert profile is not None
-        assert "personality" in profile
+        assert "disposition" in profile
         assert "background" in profile
 
-        personality = profile["personality"]
-        assert personality.openness == 0.5
-        assert personality.conscientiousness == 0.5
-        assert personality.extraversion == 0.5
-        assert personality.agreeableness == 0.5
-        assert personality.neuroticism == 0.5
-        assert personality.bias_strength == 0.5
+        disposition = profile["disposition"]
+        assert disposition.skepticism == 3
+        assert disposition.literalism == 3
+        assert disposition.empathy == 3
 
         assert profile["background"] == ""
 
     @pytest.mark.asyncio
-    async def test_update_agent_personality(self, memory: MemoryEngine):
-        """Test updating agent personality traits."""
+    async def test_update_agent_disposition(self, memory: MemoryEngine):
+        """Test updating agent disposition traits."""
         bank_id = unique_agent_id("test_profile_update")
 
         profile = await memory.get_bank_profile(bank_id)
-        assert profile["personality"].openness == 0.5
+        assert profile["disposition"].skepticism == 3
 
-        new_personality = {
-            "openness": 0.8,
-            "conscientiousness": 0.6,
-            "extraversion": 0.7,
-            "agreeableness": 0.4,
-            "neuroticism": 0.3,
-            "bias_strength": 0.9,
+        new_disposition = {
+            "skepticism": 5,
+            "literalism": 4,
+            "empathy": 2,
         }
-        await memory.update_bank_personality(bank_id, new_personality)
+        await memory.update_bank_disposition(bank_id, new_disposition)
 
         updated_profile = await memory.get_bank_profile(bank_id)
-        personality = updated_profile["personality"]
-        assert abs(personality.openness - new_personality["openness"]) < 0.001
-        assert abs(personality.conscientiousness - new_personality["conscientiousness"]) < 0.001
-        assert abs(personality.extraversion - new_personality["extraversion"]) < 0.001
-        assert abs(personality.agreeableness - new_personality["agreeableness"]) < 0.001
-        assert abs(personality.neuroticism - new_personality["neuroticism"]) < 0.001
-        assert abs(personality.bias_strength - new_personality["bias_strength"]) < 0.001
+        disposition = updated_profile["disposition"]
+        assert disposition.skepticism == new_disposition["skepticism"]
+        assert disposition.literalism == new_disposition["literalism"]
+        assert disposition.empathy == new_disposition["empathy"]
 
     @pytest.mark.asyncio
     async def test_list_agents(self, memory: MemoryEngine):
@@ -84,7 +75,7 @@ class TestAgentProfile:
 
         for agent in agents:
             assert "bank_id" in agent
-            assert "personality" in agent
+            assert "disposition" in agent
             assert "background" in agent
             assert "created_at" in agent
             assert "updated_at" in agent
@@ -104,14 +95,14 @@ class TestAgentBackground:
         result1 = await memory.merge_bank_background(
             bank_id,
             "I was born in Texas",
-            update_personality=False
+            update_disposition=False
         )
         assert "Texas" in result1["background"]
 
         result2 = await memory.merge_bank_background(
             bank_id,
             "I have 10 years of startup experience",
-            update_personality=False
+            update_disposition=False
         )
         assert "Texas" in result2["background"] or "startup" in result2["background"]
 
@@ -126,14 +117,14 @@ class TestAgentBackground:
         result1 = await memory.merge_bank_background(
             bank_id,
             "I was born in Colorado",
-            update_personality=False
+            update_disposition=False
         )
         assert "Colorado" in result1["background"]
 
         result2 = await memory.merge_bank_background(
             bank_id,
             "You were born in Texas",
-            update_personality=False
+            update_disposition=False
         )
         assert "Texas" in result2["background"]
 
@@ -147,23 +138,20 @@ class TestAgentEndpoint:
         bank_id = unique_agent_id("test_put_create")
 
         request = CreateBankRequest(
-            personality=DispositionTraits(
-                openness=0.8,
-                conscientiousness=0.6,
-                extraversion=0.5,
-                agreeableness=0.7,
-                neuroticism=0.3,
-                bias_strength=0.7
+            disposition=DispositionTraits(
+                skepticism=4,
+                literalism=5,
+                empathy=2
             ),
             background="I am a creative software engineer"
         )
 
         profile = await memory.get_bank_profile(bank_id)
 
-        if request.personality is not None:
-            await memory.update_bank_personality(
+        if request.disposition is not None:
+            await memory.update_bank_disposition(
                 bank_id,
-                request.personality.model_dump()
+                request.disposition.model_dump()
             )
 
         if request.background is not None:
@@ -182,8 +170,8 @@ class TestAgentEndpoint:
 
         final_profile = await memory.get_bank_profile(bank_id)
 
-        assert final_profile["personality"].openness == 0.8
-        assert final_profile["personality"].bias_strength == 0.7
+        assert final_profile["disposition"].skepticism == 4
+        assert final_profile["disposition"].literalism == 5
         assert final_profile["background"] == "I am a creative software engineer"
 
     @pytest.mark.asyncio
@@ -213,32 +201,29 @@ class TestAgentEndpoint:
 
         final_profile = await memory.get_bank_profile(bank_id)
 
-        assert final_profile["personality"].openness == 0.5
+        assert final_profile["disposition"].skepticism == 3  # Default
         assert final_profile["background"] == "I am a data scientist"
 
 
-class TestAgentPersonalityIntegration:
-    """Tests for personality integration with other features."""
+class TestAgentDispositionIntegration:
+    """Tests for disposition integration with other features."""
 
     @pytest.mark.asyncio
-    async def test_think_uses_personality(self, memory: MemoryEngine):
-        """Test that THINK operation uses agent personality."""
+    async def test_think_uses_disposition(self, memory: MemoryEngine):
+        """Test that THINK operation uses agent disposition."""
         bank_id = unique_agent_id("test_think")
 
-        personality = {
-            "openness": 0.9,
-            "conscientiousness": 0.2,
-            "extraversion": 0.8,
-            "agreeableness": 0.1,
-            "neuroticism": 0.7,
-            "bias_strength": 0.9,
+        disposition = {
+            "skepticism": 5,  # Very skeptical
+            "literalism": 4,  # High literalism
+            "empathy": 2,     # Low empathy
         }
-        await memory.update_bank_personality(bank_id, personality)
+        await memory.update_bank_disposition(bank_id, disposition)
 
         await memory.merge_bank_background(
             bank_id,
             "I am a creative artist who values innovation over tradition",
-            update_personality=False
+            update_disposition=False
         )
 
         await memory.retain_batch_async(
diff --git a/hindsight-api/tests/test_fact_extraction_quality.py b/hindsight-api/tests/test_fact_extraction_quality.py
index 70088346..f075baba 100644
--- a/hindsight-api/tests/test_fact_extraction_quality.py
+++ b/hindsight-api/tests/test_fact_extraction_quality.py
@@ -1009,160 +1009,162 @@ so the algorithm learns to box out. See you next week!
 
 
 # =============================================================================
-# PERSONALITY INFERENCE TESTS
+# DISPOSITION INFERENCE TESTS
 # =============================================================================
 
-class TestPersonalityInference:
-    """Tests for LLM-based personality trait inference from background."""
+class TestDispositionInference:
+    """Tests for LLM-based disposition trait inference from background."""
 
     @pytest.mark.asyncio
-    async def test_background_merge_with_personality_inference(self, memory):
-        """Test that background merge infers personality traits by default."""
+    async def test_background_merge_with_disposition_inference(self, memory):
+        """Test that background merge infers disposition traits by default."""
         import uuid
         bank_id = f"test_infer_{uuid.uuid4().hex[:8]}"
 
         result = await memory.merge_bank_background(
             bank_id,
             "I am a creative software engineer who loves innovation and trying new technologies",
-            update_personality=True
+            update_disposition=True
         )
 
         assert "background" in result
-        assert "personality" in result
+        assert "disposition" in result
 
         background = result["background"]
-        personality = result["personality"]
+        disposition = result["disposition"]
 
         assert "creative" in background.lower() or "innovation" in background.lower()
 
-        assert "openness" in personality
-        assert personality["openness"] > 0.5
-        assert 0.0 <= personality["openness"] <= 1.0
-
-        required_traits = ["openness", "conscientiousness", "extraversion",
-                          "agreeableness", "neuroticism", "bias_strength"]
+        # Check that new traits are present with valid values (1-5)
+        required_traits = ["skepticism", "literalism", "empathy"]
         for trait in required_traits:
-            assert trait in personality
-            assert 0.0 <= personality[trait] <= 1.0
+            assert trait in disposition
+            assert 1 <= disposition[trait] <= 5
 
     @pytest.mark.asyncio
-    async def test_background_merge_without_personality_inference(self, memory):
-        """Test that background merge skips personality inference when disabled."""
+    async def test_background_merge_without_disposition_inference(self, memory):
+        """Test that background merge skips disposition inference when disabled."""
         import uuid
         bank_id = f"test_no_infer_{uuid.uuid4().hex[:8]}"
 
         initial_profile = await memory.get_bank_profile(bank_id)
-        initial_personality = initial_profile["personality"]
+        initial_disposition = initial_profile["disposition"]
 
         result = await memory.merge_bank_background(
             bank_id,
             "I am a data scientist",
-            update_personality=False
+            update_disposition=False
         )
 
         assert "background" in result
-        assert "personality" not in result
+        assert "disposition" not in result
 
         final_profile = await memory.get_bank_profile(bank_id)
-        final_personality = final_profile["personality"]
+        final_disposition = final_profile["disposition"]
 
-        assert initial_personality == final_personality
+        assert initial_disposition == final_disposition
 
     @pytest.mark.asyncio
-    async def test_personality_inference_for_organized_engineer(self, memory):
-        """Test personality inference for organized/conscientious profile."""
+    async def test_disposition_inference_for_lawyer(self, memory):
+        """Test disposition inference for lawyer profile (high skepticism, high literalism)."""
         import uuid
-        bank_id = f"test_organized_{uuid.uuid4().hex[:8]}"
+        bank_id = f"test_lawyer_{uuid.uuid4().hex[:8]}"
 
         result = await memory.merge_bank_background(
             bank_id,
-            "I am a methodical engineer who values organization and systematic planning",
-            update_personality=True
+            "I am a lawyer who focuses on contract details and never takes claims at face value",
+            update_disposition=True
         )
 
-        personality = result["personality"]
+        disposition = result["disposition"]
 
-        assert personality["conscientiousness"] > 0.5
+        # Lawyers should have higher skepticism and literalism
+        assert disposition["skepticism"] >= 3
+        assert disposition["literalism"] >= 3
 
     @pytest.mark.asyncio
-    async def test_personality_inference_for_startup_founder(self, memory):
-        """Test personality inference for entrepreneurial profile."""
+    async def test_disposition_inference_for_therapist(self, memory):
+        """Test disposition inference for therapist profile (high empathy)."""
         import uuid
-        bank_id = f"test_founder_{uuid.uuid4().hex[:8]}"
+        bank_id = f"test_therapist_{uuid.uuid4().hex[:8]}"
 
         result = await memory.merge_bank_background(
             bank_id,
-            "I am a startup founder who thrives on risk and social interaction",
-            update_personality=True
+            "I am a therapist who deeply understands and connects with people's emotional struggles",
+            update_disposition=True
         )
 
-        personality = result["personality"]
+        disposition = result["disposition"]
 
-        assert personality["openness"] > 0.5
-        assert personality["extraversion"] > 0.5
+        # Therapists should have higher empathy
+        assert disposition["empathy"] >= 3
 
     @pytest.mark.asyncio
-    async def test_personality_updates_in_database(self, memory):
-        """Test that inferred personality is actually stored in database."""
+    async def test_disposition_updates_in_database(self, memory):
+        """Test that inferred disposition is actually stored in database."""
         import uuid
         bank_id = f"test_db_update_{uuid.uuid4().hex[:8]}"
 
         result = await memory.merge_bank_background(
             bank_id,
             "I am an innovative designer",
-            update_personality=True
+            update_disposition=True
         )
 
-        inferred_personality = result["personality"]
+        inferred_disposition = result["disposition"]
 
         profile = await memory.get_bank_profile(bank_id)
-        db_personality = profile["personality"]
+        db_disposition = profile["disposition"]
 
-        assert db_personality == inferred_personality
+        # Compare values (db_disposition is a Pydantic model)
+        assert db_disposition.skepticism == inferred_disposition["skepticism"]
+        assert db_disposition.literalism == inferred_disposition["literalism"]
+        assert db_disposition.empathy == inferred_disposition["empathy"]
 
     @pytest.mark.asyncio
-    async def test_multiple_background_merges_update_personality(self, memory):
-        """Test that each background merge can update personality."""
+    async def test_multiple_background_merges_update_disposition(self, memory):
+        """Test that each background merge can update disposition."""
         import uuid
         bank_id = f"test_multi_merge_{uuid.uuid4().hex[:8]}"
 
         result1 = await memory.merge_bank_background(
             bank_id,
             "I am a software engineer",
-            update_personality=True
+            update_disposition=True
         )
-        personality1 = result1["personality"]
+        disposition1 = result1["disposition"]
 
         result2 = await memory.merge_bank_background(
             bank_id,
             "I love creative problem solving and innovation",
-            update_personality=True
+            update_disposition=True
         )
-        personality2 = result2["personality"]
+        disposition2 = result2["disposition"]
 
         assert "engineer" in result2["background"].lower() or "software" in result2["background"].lower()
         assert "creative" in result2["background"].lower() or "innovation" in result2["background"].lower()
 
     @pytest.mark.asyncio
-    async def test_background_merge_conflict_resolution_with_personality(self, memory):
-        """Test that conflicts are resolved and personality reflects final background."""
+    async def test_background_merge_conflict_resolution_with_disposition(self, memory):
+        """Test that conflicts are resolved and disposition reflects final background."""
         import uuid
         bank_id = f"test_conflict_{uuid.uuid4().hex[:8]}"
 
         await memory.merge_bank_background(
             bank_id,
             "I was born in Colorado and prefer stability",
-            update_personality=True
+            update_disposition=True
         )
 
         result = await memory.merge_bank_background(
             bank_id,
-            "You were born in Texas and love taking risks",
-            update_personality=True
+            "You were born in Texas and are very skeptical of people",
+            update_disposition=True
         )
 
         background = result["background"]
-        personality = result["personality"]
+        disposition = result["disposition"]
 
         assert "texas" in background.lower()
-        assert personality["openness"] > 0.5
\ No newline at end of file
+        # Higher skepticism expected from "very skeptical of people"
+        assert disposition["skepticism"] >= 3
\ No newline at end of file
diff --git a/hindsight-api/tests/test_fact_ordering.py b/hindsight-api/tests/test_fact_ordering.py
index 33147067..ef4aceb8 100644
--- a/hindsight-api/tests/test_fact_ordering.py
+++ b/hindsight-api/tests/test_fact_ordering.py
@@ -19,14 +19,11 @@ async def test_fact_ordering_within_conversation(memory):
     # Get/create agent (auto-creates with defaults)
     await memory.get_bank_profile(bank_id)
 
-    # Update personality to match Marcus
-    await memory.update_bank_personality(bank_id, {
-        "openness": 0.7,
-        "conscientiousness": 0.6,
-        "extraversion": 0.8,
-        "agreeableness": 0.5,
-        "neuroticism": 0.3,
-        "bias_strength": 0.5
+    # Update disposition to match Marcus
+    await memory.update_bank_disposition(bank_id, {
+        "skepticism": 3,
+        "literalism": 3,
+        "empathy": 3
     })
 
     # A conversation where Marcus changes his position
diff --git a/hindsight-api/tests/test_http_api_integration.py b/hindsight-api/tests/test_http_api_integration.py
index 7dce716e..df305740 100644
--- a/hindsight-api/tests/test_http_api_integration.py
+++ b/hindsight-api/tests/test_http_api_integration.py
@@ -60,9 +60,9 @@ async def test_full_api_workflow(api_client, test_bank_id):
     response = await api_client.get(f"/v1/default/banks/{test_bank_id}/profile")
     assert response.status_code == 200
     profile = response.json()
-    assert "personality" in profile
+    assert "disposition" in profile
     assert "background" in profile
-    print(f"Bank profile created with personality: {profile['personality']}")
+    print(f"Bank profile created with disposition: {profile['disposition']}")
 
     # Add background
     response = await api_client.post(
@@ -237,27 +237,24 @@ async def test_full_api_workflow(api_client, test_bank_id):
     # Note: Document deletion is tested separately in test_document_deletion
 
     # ================================================================
-    # 7. Update and Verify Bank Personality
+    # 7. Update and Verify Bank Disposition
     # ================================================================
 
-    # Update personality traits
+    # Update disposition traits
     response = await api_client.put(
         f"/v1/default/banks/{test_bank_id}/profile",
         json={
-            "personality": {
-                "openness": 0.8,
-                "conscientiousness": 0.7,
-                "extraversion": 0.6,
-                "agreeableness": 0.9,
-                "neuroticism": 0.3,
-                "bias_strength": 0.5
+            "disposition": {
+                "skepticism": 4,
+                "literalism": 3,
+                "empathy": 4
             }
         }
     )
     assert response.status_code == 200
-    print("Personality updated")
+    print("Disposition updated")
 
-    # Check profile again (should have updated personality)
+    # Check profile again (should have updated disposition)
     response = await api_client.get(f"/v1/default/banks/{test_bank_id}/profile")
     assert response.status_code == 200
     updated_profile = response.json()
diff --git a/hindsight-api/tests/test_observations.py b/hindsight-api/tests/test_observations.py
index 11c55260..1085d036 100644
--- a/hindsight-api/tests/test_observations.py
+++ b/hindsight-api/tests/test_observations.py
@@ -9,32 +9,39 @@ from datetime import datetime, timezone
 @pytest.mark.asyncio
 async def test_observation_generation_on_put(memory):
     """
-    Test that observations are generated when new facts are added.
+    Test that observations are generated SYNCHRONOUSLY when new facts are added.
 
-    1. Store facts about an entity
-    2. Wait for background tasks (observation generation)
-    3. Verify observations were created and linked to the entity
+    Observations are generated during retain when:
+    - Entity has >= 5 facts (MIN_FACTS_THRESHOLD)
+    - Entity is in top 5 by mention count
+
+    This test stores enough facts to trigger automatic observation generation.
     """
     bank_id = f"test_obs_{datetime.now(timezone.utc).timestamp()}"
 
     try:
-        # Store some facts about an entity
-        await memory.retain_async(
-            bank_id=bank_id,
-            content="John is a software engineer at Google. He is detail-oriented and methodical.",
-            context="work info",
-            event_date=datetime(2024, 1, 15, tzinfo=timezone.utc)
-        )
+        # Store multiple facts about John to reach the MIN_FACTS_THRESHOLD (5)
+        # Each retain call should extract at least one fact about John
+        contents = [
+            "John is a software engineer at Google.",
+            "John is detail-oriented and methodical in his work.",
+            "John has been working on the AI team for 3 years.",
+            "John specializes in machine learning and deep learning.",
+            "John presented at the company conference last week.",
+            "John mentors junior engineers on the team.",
+        ]
 
-        await memory.retain_async(
-            bank_id=bank_id,
-            content="John has been working on the AI team for 3 years. He specializes in machine learning.",
-            context="work info",
-            event_date=datetime(2024, 2, 1, tzinfo=timezone.utc)
-        )
+        for i, content in enumerate(contents):
+            await memory.retain_async(
+                bank_id=bank_id,
+                content=content,
+                context="work info",
+                event_date=datetime(2024, 1, 15 + i, tzinfo=timezone.utc)
+            )
 
-        # Wait for background tasks to complete (including observation generation)
-        await memory.wait_for_background_tasks()
+        # Observations are generated SYNCHRONOUSLY during retain,
+        # so they should be available immediately after retain completes.
+        # No need to wait for background tasks for observations.
 
         # Find the John entity
         pool = await memory._get_pool()
@@ -49,32 +56,42 @@ async def test_observation_generation_on_put(memory):
                 bank_id
             )
 
-        if entity_row:
-            entity_id = str(entity_row['id'])
-            entity_name = entity_row['canonical_name']
-            print(f"\n=== Found Entity ===")
-            print(f"Entity: {entity_name} (id: {entity_id})")
+            # Also check the fact count for this entity
+            if entity_row:
+                fact_count = await conn.fetchval(
+                    """
+                    SELECT COUNT(*) FROM unit_entities WHERE entity_id = $1
+                    """,
+                    entity_row['id']
+                )
+                print(f"\n=== Entity Facts ===")
+                print(f"Entity: {entity_row['canonical_name']} has {fact_count} linked facts")
 
-            # Get observations for the entity
-            observations = await memory.get_entity_observations(bank_id, entity_id, limit=10)
+        assert entity_row is not None, "John entity should have been extracted"
 
-            print(f"\n=== Observations for {entity_name} ===")
-            print(f"Total observations: {len(observations)}")
-            for obs in observations:
-                print(f"  - {obs.text}")
+        entity_id = str(entity_row['id'])
+        entity_name = entity_row['canonical_name']
+        print(f"\n=== Found Entity ===")
+        print(f"Entity: {entity_name} (id: {entity_id})")
 
-            # Verify observations were created
-            if len(observations) > 0:
-                print(f"✓ Observations were successfully generated")
-                # Check that observations mention relevant content
-                obs_texts = " ".join([o.text.lower() for o in observations])
-                assert any(keyword in obs_texts for keyword in ["google", "engineer", "ai", "machine learning", "detail"]), \
-                    "Observations should contain relevant information about John"
-            else:
-                print(f"⚠ Note: No observations were generated (this can happen if LLM extraction varies)")
+        # Get observations for the entity - should be available immediately
+        observations = await memory.get_entity_observations(bank_id, entity_id, limit=10)
 
-        else:
-            print(f"⚠ Note: No 'John' entity was extracted (LLM extraction may vary)")
+        print(f"\n=== Observations for {entity_name} ===")
+        print(f"Total observations: {len(observations)}")
+        for obs in observations:
+            print(f"  - {obs.text}")
+
+        # Verify observations were created (requires >= 5 facts)
+        assert len(observations) > 0, \
+            f"Observations should have been generated synchronously during retain (entity has {fact_count} facts, threshold is 5)"
+
+        # Check that observations mention relevant content
+        obs_texts = " ".join([o.text.lower() for o in observations])
+        assert any(keyword in obs_texts for keyword in ["google", "engineer", "ai", "machine learning", "detail"]), \
+            "Observations should contain relevant information about John"
+
+        print(f"✓ Observations were successfully generated synchronously during retain")
 
     finally:
         # Cleanup
@@ -156,34 +173,40 @@ async def test_regenerate_entity_observations(memory):
 async def test_search_with_include_entities(memory):
     """
     Test that search with include_entities=True returns entity observations.
+
+    This test verifies that:
+    1. Observations are generated during retain (when entity has >= 5 facts)
+    2. Observations are returned in recall results with include_entities=True
     """
     bank_id = f"test_search_ent_{datetime.now(timezone.utc).timestamp()}"
 
     try:
-        # Store facts about entities
-        await memory.retain_async(
-            bank_id=bank_id,
-            content="Alice is a data scientist who works on recommendation systems at Netflix.",
-            context="work info",
-            event_date=datetime(2024, 1, 15, tzinfo=timezone.utc)
-        )
+        # Store enough facts about Alice to trigger observation generation (>= 5 facts)
+        contents = [
+            "Alice is a data scientist who works on recommendation systems at Netflix.",
+            "Alice presented her research at the ML conference last month.",
+            "Alice is an expert in deep learning and neural networks.",
+            "Alice graduated from Stanford with a PhD in Computer Science.",
+            "Alice leads a team of 5 data scientists at Netflix.",
+            "Alice published a paper on collaborative filtering algorithms.",
+        ]
 
-        await memory.retain_async(
-            bank_id=bank_id,
-            content="Alice presented her research at the ML conference last month. She is an expert in deep learning.",
-            context="work info",
-            event_date=datetime(2024, 2, 1, tzinfo=timezone.utc)
-        )
+        for i, content in enumerate(contents):
+            await memory.retain_async(
+                bank_id=bank_id,
+                content=content,
+                context="work info",
+                event_date=datetime(2024, 1, 15 + i, tzinfo=timezone.utc)
+            )
 
-        # Wait for background tasks
-        await memory.wait_for_background_tasks()
+        # Observations are generated synchronously during retain, no need to wait
 
         # Search with include_entities=True
         result = await memory.recall_async(
             bank_id=bank_id,
             query="What does Alice do?",
-            fact_type=["world", "agent"],
-            budget=Budget.LOW, # 30,
+            fact_type=["world", "experience"],
+            budget=Budget.LOW,
             max_tokens=2000,
             include_entities=True,
             max_entity_tokens=500
@@ -196,7 +219,7 @@ async def test_search_with_include_entities(memory):
             if fact.entities:
                 print(f"    Entities: {', '.join(fact.entities)}")
 
-        print(f"\n=== Entity Observations ===")
+        print(f"\n=== Entity Observations in Recall ===")
         if result.entities:
             for name, state in result.entities.items():
                 print(f"\n{name}:")
@@ -210,15 +233,26 @@ async def test_search_with_include_entities(memory):
 
         # Check if entities are included in facts
         facts_with_entities = [f for f in result.results if f.entities]
-        if facts_with_entities:
-            print(f"✓ {len(facts_with_entities)} facts have entity information")
+        assert len(facts_with_entities) > 0, "Some facts should have entity information"
+        print(f"✓ {len(facts_with_entities)} facts have entity information")
 
-        # Check if entity observations are included
-        if result.entities:
-            print(f"✓ Entity observations included for {len(result.entities)} entities")
-            for name, state in result.entities.items():
-                assert state.canonical_name == name, "Entity canonical_name should match key"
-                assert state.entity_id, "Entity should have an ID"
+        # Check if entity observations are included in recall
+        assert result.entities is not None and len(result.entities) > 0, \
+            "Entity observations should be included in recall results"
+        print(f"✓ Entity observations included for {len(result.entities)} entities")
+
+        # Verify Alice entity has observations
+        alice_found = False
+        for name, state in result.entities.items():
+            assert state.canonical_name == name, "Entity canonical_name should match key"
+            assert state.entity_id, "Entity should have an ID"
+            if "alice" in name.lower():
+                alice_found = True
+                assert len(state.observations) > 0, \
+                    "Alice should have observations (generated during retain)"
+                print(f"✓ Alice has {len(state.observations)} observations in recall result")
+
+        assert alice_found, "Alice entity should be in recall results"
 
     finally:
         # Cleanup
@@ -337,3 +371,127 @@ async def test_observation_fact_type_in_database(memory):
         async with pool.acquire() as conn:
             await conn.execute("DELETE FROM memory_units WHERE bank_id = $1", bank_id)
             await conn.execute("DELETE FROM entities WHERE bank_id = $1", bank_id)
+
+
+@pytest.mark.asyncio
+async def test_user_entity_prioritized_for_observations(memory):
+    """
+    Test that the 'user' entity gets observations even when many other entities exist.
+
+    The retain pipeline only regenerates observations for TOP_N_ENTITIES (5) entities,
+    sorted by mention count. This test verifies that the most mentioned entity ('user')
+    gets prioritized and receives observations.
+
+    This is critical because 'user' is often the most important entity in personal memory.
+    """
+    bank_id = f"test_user_priority_{datetime.now(timezone.utc).timestamp()}"
+
+    try:
+        # Create content where 'user' (the user) is mentioned many times
+        # along with several other entities
+        contents = [
+            # User mentioned frequently
+            "The user loves hiking in the mountains during summer.",
+            "The user works as a software engineer at Microsoft.",
+            "The user has a dog named Max who is a golden retriever.",
+            "The user enjoys cooking Italian food, especially pasta.",
+            "The user graduated from MIT with a Computer Science degree.",
+            "The user's favorite book is 'Dune' by Frank Herbert.",
+            # Other entities mentioned fewer times
+            "Sarah is a friend who works at Google.",
+            "Bob is a colleague from the data science team.",
+            "Tokyo is a city the user visited last year.",
+            "Python is the user's favorite programming language.",
+        ]
+
+        # Retain all content in a single batch for efficiency
+        for i, content in enumerate(contents):
+            await memory.retain_async(
+                bank_id=bank_id,
+                content=content,
+                context="personal info",
+                event_date=datetime(2024, 1, 15 + i, tzinfo=timezone.utc)
+            )
+
+        # Observations are generated synchronously during retain
+
+        # Find the 'user' entity
+        pool = await memory._get_pool()
+        async with pool.acquire() as conn:
+            # Find user entity (may be named "user", "the user", etc.)
+            user_entity = await conn.fetchrow(
+                """
+                SELECT e.id, e.canonical_name,
+                       (SELECT COUNT(*) FROM unit_entities ue
+                        JOIN memory_units mu ON ue.unit_id = mu.id
+                        WHERE ue.entity_id = e.id AND mu.bank_id = $1) as fact_count
+                FROM entities e
+                WHERE e.bank_id = $1
+                  AND LOWER(e.canonical_name) LIKE '%user%'
+                LIMIT 1
+                """,
+                bank_id
+            )
+
+            # Get all entities with their fact counts to verify prioritization
+            all_entities = await conn.fetch(
+                """
+                SELECT e.id, e.canonical_name,
+                       (SELECT COUNT(*) FROM unit_entities ue
+                        JOIN memory_units mu ON ue.unit_id = mu.id
+                        WHERE ue.entity_id = e.id AND mu.bank_id = $1) as fact_count
+                FROM entities e
+                WHERE e.bank_id = $1
+                ORDER BY fact_count DESC
+                """,
+                bank_id
+            )
+
+        print(f"\n=== Entities by Mention Count ===")
+        for entity in all_entities:
+            print(f"  {entity['canonical_name']}: {entity['fact_count']} mentions")
+
+        # Verify user entity exists
+        assert user_entity is not None, "User entity should have been extracted"
+        user_entity_id = str(user_entity['id'])
+        user_entity_name = user_entity['canonical_name']
+        user_fact_count = user_entity['fact_count']
+
+        print(f"\n=== User Entity ===")
+        print(f"Entity: {user_entity_name} (id: {user_entity_id})")
+        print(f"Fact count: {user_fact_count}")
+
+        # Verify user has enough facts for observations (>= MIN_FACTS_THRESHOLD of 5)
+        assert user_fact_count >= 5, \
+            f"User entity should have at least 5 facts, but has {user_fact_count}"
+
+        # Get observations for user entity
+        observations = await memory.get_entity_observations(bank_id, user_entity_id, limit=10)
+
+        print(f"\n=== User Entity Observations ===")
+        print(f"Total observations: {len(observations)}")
+        for obs in observations:
+            print(f"  - {obs.text}")
+
+        # Verify observations were generated for user (critical assertion)
+        assert len(observations) > 0, \
+            f"User entity should have observations (has {user_fact_count} facts, threshold is 5). " \
+            f"This may indicate that 'user' is not being prioritized in the top 5 entities by mention count."
+
+        # Verify observations mention relevant content about the user
+        obs_texts = " ".join([o.text.lower() for o in observations])
+        user_keywords = ["hiking", "software", "engineer", "dog", "max", "cooking",
+                        "italian", "mit", "dune", "microsoft"]
+        matching_keywords = [k for k in user_keywords if k in obs_texts]
+        assert len(matching_keywords) > 0, \
+            f"Observations should contain relevant information about the user. Keywords found: {matching_keywords}"
+
+        print(f"✓ User entity was prioritized and received {len(observations)} observations")
+        print(f"✓ Observations contain relevant keywords: {matching_keywords}")
+
+    finally:
+        # Cleanup
+        pool = await memory._get_pool()
+        async with pool.acquire() as conn:
+            await conn.execute("DELETE FROM memory_units WHERE bank_id = $1", bank_id)
+            await conn.execute("DELETE FROM entities WHERE bank_id = $1", bank_id)
diff --git a/hindsight-cli/src/commands/bank.rs b/hindsight-cli/src/commands/bank.rs
index 84ed1926..9d63016f 100644
--- a/hindsight-cli/src/commands/bank.rs
+++ b/hindsight-cli/src/commands/bank.rs
@@ -231,11 +231,9 @@ pub fn update_background(
                         (current_profile.as_ref().map(|p| p.disposition.clone()), &profile.disposition)
                     {
                         println!("\nDisposition changes:");
-                        println!("  Openness:          {:.2} → {:.2}", old_p.openness, new_p.openness);
-                        println!("  Conscientiousness: {:.2} → {:.2}", old_p.conscientiousness, new_p.conscientiousness);
-                        println!("  Extraversion:      {:.2} → {:.2}", old_p.extraversion, new_p.extraversion);
-                        println!("  Agreeableness:     {:.2} → {:.2}", old_p.agreeableness, new_p.agreeableness);
-                        println!("  Neuroticism:       {:.2} → {:.2}", old_p.neuroticism, new_p.neuroticism);
+                        println!("  Skepticism:  {} → {}", old_p.skepticism, new_p.skepticism);
+                        println!("  Literalism:  {} → {}", old_p.literalism, new_p.literalism);
+                        println!("  Empathy:     {} → {}", old_p.empathy, new_p.empathy);
                     }
                 }
             } else {
diff --git a/hindsight-cli/src/ui.rs b/hindsight-cli/src/ui.rs
index 34879c61..1a78cf26 100644
--- a/hindsight-cli/src/ui.rs
+++ b/hindsight-cli/src/ui.rs
@@ -209,17 +209,17 @@ pub fn print_profile(profile: &BankProfileResponse) {
     println!("{}", "─── Disposition Traits ───".bright_yellow());
     println!();
 
-    let traits = [
-        ("Openness", profile.disposition.openness, "🔓", "green"),
-        ("Conscientiousness", profile.disposition.conscientiousness, "📋", "yellow"),
-        ("Extraversion", profile.disposition.extraversion, "🗣️", "cyan"),
-        ("Agreeableness", profile.disposition.agreeableness, "🤝", "magenta"),
-        ("Neuroticism", profile.disposition.neuroticism, "😰", "yellow"),
+    // New 3-trait disposition system (values 1-5)
+    let traits: [(_, i64, _, _, _); 3] = [
+        ("Skepticism", profile.disposition.skepticism, "🔍", "cyan", "1=trusting, 5=skeptical"),
+        ("Literalism", profile.disposition.literalism, "📋", "yellow", "1=flexible, 5=literal"),
+        ("Empathy", profile.disposition.empathy, "💚", "green", "1=detached, 5=empathetic"),
     ];
 
-    for (name, value, emoji, color) in &traits {
+    for (name, value, emoji, color, desc) in &traits {
+        // Scale 1-5 to bar visualization (each point = 8 chars, total 40)
         let bar_length = 40;
-        let filled = (*value * bar_length as f64) as usize;
+        let filled = ((*value - 1) * 10) as usize; // 1->0, 2->10, 3->20, 4->30, 5->40
         let empty = bar_length - filled;
 
         let bar = format!("{}{}", "█".repeat(filled), "░".repeat(empty));
@@ -231,27 +231,14 @@ pub fn print_profile(profile: &BankProfileResponse) {
             _ => bar.bright_white(),
         };
 
-        println!("  {} {:<20} [{}] {:.0}%",
+        println!("  {} {:<12} [{}] {}/5",
             emoji,
             name,
             colored_bar,
-            value * 100.0
+            value
         );
+        println!("    {}", desc.bright_black());
     }
 
     println!();
-    println!("{}", "Bias Strength:".bright_yellow());
-    let bias = profile.disposition.bias_strength;
-    let bar_length = 40;
-    let filled = (bias * bar_length as f64) as usize;
-    let empty = bar_length - filled;
-    let bar = format!("{}{}", "█".repeat(filled), "░".repeat(empty));
-
-    println!("  💪 {:<20} [{}] {:.0}%",
-        "Disposition Influence",
-        bar.bright_green(),
-        bias * 100.0
-    );
-    println!("  {}", "(how much disposition shapes opinions)".bright_black());
-    println!();
 }
diff --git a/hindsight-clients/python/hindsight_client_api/api/default_api.py b/hindsight-clients/python/hindsight_client_api/api/default_api.py
index 4e7c7a07..e6f212d1 100644
--- a/hindsight-clients/python/hindsight_client_api/api/default_api.py
+++ b/hindsight-clients/python/hindsight_client_api/api/default_api.py
@@ -5705,7 +5705,7 @@ class DefaultApi:
     ) -> BankProfileResponse:
         """Update memory bank disposition
 
-        Update bank's Big Five disposition traits and bias strength
+        Update bank's disposition traits (skepticism, literalism, empathy)
 
         :param bank_id: (required)
         :type bank_id: str
@@ -5777,7 +5777,7 @@ class DefaultApi:
     ) -> ApiResponse[BankProfileResponse]:
         """Update memory bank disposition
 
-        Update bank's Big Five disposition traits and bias strength
+        Update bank's disposition traits (skepticism, literalism, empathy)
 
         :param bank_id: (required)
         :type bank_id: str
@@ -5849,7 +5849,7 @@ class DefaultApi:
     ) -> RESTResponseType:
         """Update memory bank disposition
 
-        Update bank's Big Five disposition traits and bias strength
+        Update bank's disposition traits (skepticism, literalism, empathy)
 
         :param bank_id: (required)
         :type bank_id: str
diff --git a/hindsight-clients/python/hindsight_client_api/docs/AddBackgroundRequest.md b/hindsight-clients/python/hindsight_client_api/docs/AddBackgroundRequest.md
index 0e4d4ca0..3f000fd2 100644
--- a/hindsight-clients/python/hindsight_client_api/docs/AddBackgroundRequest.md
+++ b/hindsight-clients/python/hindsight_client_api/docs/AddBackgroundRequest.md
@@ -7,7 +7,7 @@ Request model for adding/merging background information.
 Name | Type | Description | Notes
 ------------ | ------------- | ------------- | -------------
 **content** | **str** | New background information to add or merge | 
-**update_disposition** | **bool** | If true, infer Big Five disposition traits from the merged background (default: true) | [optional] [default to True]
+**update_disposition** | **bool** | If true, infer disposition traits from the merged background (default: true) | [optional] [default to True]
 
 ## Example
 
diff --git a/hindsight-clients/python/hindsight_client_api/docs/DefaultApi.md b/hindsight-clients/python/hindsight_client_api/docs/DefaultApi.md
index d2fffde4..99b77bc5 100644
--- a/hindsight-clients/python/hindsight_client_api/docs/DefaultApi.md
+++ b/hindsight-clients/python/hindsight_client_api/docs/DefaultApi.md
@@ -1499,7 +1499,7 @@ No authorization required
 
 Update memory bank disposition
 
-Update bank's Big Five disposition traits and bias strength
+Update bank's disposition traits (skepticism, literalism, empathy)
 
 ### Example
 
diff --git a/hindsight-clients/python/hindsight_client_api/docs/DispositionTraits.md b/hindsight-clients/python/hindsight_client_api/docs/DispositionTraits.md
index 782bb97f..4996890c 100644
--- a/hindsight-clients/python/hindsight_client_api/docs/DispositionTraits.md
+++ b/hindsight-clients/python/hindsight_client_api/docs/DispositionTraits.md
@@ -1,17 +1,14 @@
 # DispositionTraits
 
-Disposition traits based on Big Five model.
+Disposition traits that influence how memories are formed and interpreted.
 
 ## Properties
 
 Name | Type | Description | Notes
 ------------ | ------------- | ------------- | -------------
-**openness** | **float** | Openness to experience (0-1) | 
-**conscientiousness** | **float** | Conscientiousness (0-1) | 
-**extraversion** | **float** | Extraversion (0-1) | 
-**agreeableness** | **float** | Agreeableness (0-1) | 
-**neuroticism** | **float** | Neuroticism (0-1) | 
-**bias_strength** | **float** | How strongly disposition influences opinions (0-1) | 
+**skepticism** | **int** | How skeptical vs trusting (1&#x3D;trusting, 5&#x3D;skeptical) | 
+**literalism** | **int** | How literally to interpret information (1&#x3D;flexible, 5&#x3D;literal) | 
+**empathy** | **int** | How much to consider emotional context (1&#x3D;detached, 5&#x3D;empathetic) | 
 
 ## Example
 
diff --git a/hindsight-clients/python/hindsight_client_api/models/add_background_request.py b/hindsight-clients/python/hindsight_client_api/models/add_background_request.py
index f8379d31..67cd67b3 100644
--- a/hindsight-clients/python/hindsight_client_api/models/add_background_request.py
+++ b/hindsight-clients/python/hindsight_client_api/models/add_background_request.py
@@ -27,7 +27,7 @@ class AddBackgroundRequest(BaseModel):
     Request model for adding/merging background information.
     """ # noqa: E501
     content: StrictStr = Field(description="New background information to add or merge")
-    update_disposition: Optional[StrictBool] = Field(default=True, description="If true, infer Big Five disposition traits from the merged background (default: true)")
+    update_disposition: Optional[StrictBool] = Field(default=True, description="If true, infer disposition traits from the merged background (default: true)")
     __properties: ClassVar[List[str]] = ["content", "update_disposition"]
 
     model_config = ConfigDict(
diff --git a/hindsight-clients/python/hindsight_client_api/models/disposition_traits.py b/hindsight-clients/python/hindsight_client_api/models/disposition_traits.py
index d6a8dfff..cf858051 100644
--- a/hindsight-clients/python/hindsight_client_api/models/disposition_traits.py
+++ b/hindsight-clients/python/hindsight_client_api/models/disposition_traits.py
@@ -18,22 +18,19 @@ import re  # noqa: F401
 import json
 
 from pydantic import BaseModel, ConfigDict, Field
-from typing import Any, ClassVar, Dict, List, Union
+from typing import Any, ClassVar, Dict, List
 from typing_extensions import Annotated
 from typing import Optional, Set
 from typing_extensions import Self
 
 class DispositionTraits(BaseModel):
     """
-    Disposition traits based on Big Five model.
+    Disposition traits that influence how memories are formed and interpreted.
     """ # noqa: E501
-    openness: Union[Annotated[float, Field(le=1.0, strict=True, ge=0.0)], Annotated[int, Field(le=1, strict=True, ge=0)]] = Field(description="Openness to experience (0-1)")
-    conscientiousness: Union[Annotated[float, Field(le=1.0, strict=True, ge=0.0)], Annotated[int, Field(le=1, strict=True, ge=0)]] = Field(description="Conscientiousness (0-1)")
-    extraversion: Union[Annotated[float, Field(le=1.0, strict=True, ge=0.0)], Annotated[int, Field(le=1, strict=True, ge=0)]] = Field(description="Extraversion (0-1)")
-    agreeableness: Union[Annotated[float, Field(le=1.0, strict=True, ge=0.0)], Annotated[int, Field(le=1, strict=True, ge=0)]] = Field(description="Agreeableness (0-1)")
-    neuroticism: Union[Annotated[float, Field(le=1.0, strict=True, ge=0.0)], Annotated[int, Field(le=1, strict=True, ge=0)]] = Field(description="Neuroticism (0-1)")
-    bias_strength: Union[Annotated[float, Field(le=1.0, strict=True, ge=0.0)], Annotated[int, Field(le=1, strict=True, ge=0)]] = Field(description="How strongly disposition influences opinions (0-1)")
-    __properties: ClassVar[List[str]] = ["openness", "conscientiousness", "extraversion", "agreeableness", "neuroticism", "bias_strength"]
+    skepticism: Annotated[int, Field(le=5, strict=True, ge=1)] = Field(description="How skeptical vs trusting (1=trusting, 5=skeptical)")
+    literalism: Annotated[int, Field(le=5, strict=True, ge=1)] = Field(description="How literally to interpret information (1=flexible, 5=literal)")
+    empathy: Annotated[int, Field(le=5, strict=True, ge=1)] = Field(description="How much to consider emotional context (1=detached, 5=empathetic)")
+    __properties: ClassVar[List[str]] = ["skepticism", "literalism", "empathy"]
 
     model_config = ConfigDict(
         populate_by_name=True,
@@ -86,12 +83,9 @@ class DispositionTraits(BaseModel):
             return cls.model_validate(obj)
 
         _obj = cls.model_validate({
-            "openness": obj.get("openness"),
-            "conscientiousness": obj.get("conscientiousness"),
-            "extraversion": obj.get("extraversion"),
-            "agreeableness": obj.get("agreeableness"),
-            "neuroticism": obj.get("neuroticism"),
-            "bias_strength": obj.get("bias_strength")
+            "skepticism": obj.get("skepticism"),
+            "literalism": obj.get("literalism"),
+            "empathy": obj.get("empathy")
         })
         return _obj
 
diff --git a/hindsight-clients/python/hindsight_client_api/test/test_background_response.py b/hindsight-clients/python/hindsight_client_api/test/test_background_response.py
index 0c70b134..d48a57de 100644
--- a/hindsight-clients/python/hindsight_client_api/test/test_background_response.py
+++ b/hindsight-clients/python/hindsight_client_api/test/test_background_response.py
@@ -36,7 +36,7 @@ class TestBackgroundResponse(unittest.TestCase):
         if include_optional:
             return BackgroundResponse(
                 background = '',
-                disposition = {agreeableness=0.7, bias_strength=0.7, conscientiousness=0.6, extraversion=0.5, neuroticism=0.3, openness=0.8}
+                disposition = {empathy=3, literalism=3, skepticism=3}
             )
         else:
             return BackgroundResponse(
diff --git a/hindsight-clients/python/hindsight_client_api/test/test_bank_list_item.py b/hindsight-clients/python/hindsight_client_api/test/test_bank_list_item.py
index f4b7e855..a8b0a4f3 100644
--- a/hindsight-clients/python/hindsight_client_api/test/test_bank_list_item.py
+++ b/hindsight-clients/python/hindsight_client_api/test/test_bank_list_item.py
@@ -37,7 +37,7 @@ class TestBankListItem(unittest.TestCase):
             return BankListItem(
                 bank_id = '',
                 name = '',
-                disposition = {agreeableness=0.7, bias_strength=0.7, conscientiousness=0.6, extraversion=0.5, neuroticism=0.3, openness=0.8},
+                disposition = {empathy=3, literalism=3, skepticism=3},
                 background = '',
                 created_at = '',
                 updated_at = ''
@@ -46,7 +46,7 @@ class TestBankListItem(unittest.TestCase):
             return BankListItem(
                 bank_id = '',
                 name = '',
-                disposition = {agreeableness=0.7, bias_strength=0.7, conscientiousness=0.6, extraversion=0.5, neuroticism=0.3, openness=0.8},
+                disposition = {empathy=3, literalism=3, skepticism=3},
                 background = '',
         )
         """
diff --git a/hindsight-clients/python/hindsight_client_api/test/test_bank_list_response.py b/hindsight-clients/python/hindsight_client_api/test/test_bank_list_response.py
index 1989295d..3b7fc4b8 100644
--- a/hindsight-clients/python/hindsight_client_api/test/test_bank_list_response.py
+++ b/hindsight-clients/python/hindsight_client_api/test/test_bank_list_response.py
@@ -39,7 +39,7 @@ class TestBankListResponse(unittest.TestCase):
                     hindsight_client_api.models.bank_list_item.BankListItem(
                         bank_id = '', 
                         name = '', 
-                        disposition = {agreeableness=0.7, bias_strength=0.7, conscientiousness=0.6, extraversion=0.5, neuroticism=0.3, openness=0.8}, 
+                        disposition = {empathy=3, literalism=3, skepticism=3}, 
                         background = '', 
                         created_at = '', 
                         updated_at = '', )
@@ -51,7 +51,7 @@ class TestBankListResponse(unittest.TestCase):
                     hindsight_client_api.models.bank_list_item.BankListItem(
                         bank_id = '', 
                         name = '', 
-                        disposition = {agreeableness=0.7, bias_strength=0.7, conscientiousness=0.6, extraversion=0.5, neuroticism=0.3, openness=0.8}, 
+                        disposition = {empathy=3, literalism=3, skepticism=3}, 
                         background = '', 
                         created_at = '', 
                         updated_at = '', )
diff --git a/hindsight-clients/python/hindsight_client_api/test/test_bank_profile_response.py b/hindsight-clients/python/hindsight_client_api/test/test_bank_profile_response.py
index 8201111a..8874401d 100644
--- a/hindsight-clients/python/hindsight_client_api/test/test_bank_profile_response.py
+++ b/hindsight-clients/python/hindsight_client_api/test/test_bank_profile_response.py
@@ -37,14 +37,14 @@ class TestBankProfileResponse(unittest.TestCase):
             return BankProfileResponse(
                 bank_id = '',
                 name = '',
-                disposition = {agreeableness=0.7, bias_strength=0.7, conscientiousness=0.6, extraversion=0.5, neuroticism=0.3, openness=0.8},
+                disposition = {empathy=3, literalism=3, skepticism=3},
                 background = ''
             )
         else:
             return BankProfileResponse(
                 bank_id = '',
                 name = '',
-                disposition = {agreeableness=0.7, bias_strength=0.7, conscientiousness=0.6, extraversion=0.5, neuroticism=0.3, openness=0.8},
+                disposition = {empathy=3, literalism=3, skepticism=3},
                 background = '',
         )
         """
diff --git a/hindsight-clients/python/hindsight_client_api/test/test_create_bank_request.py b/hindsight-clients/python/hindsight_client_api/test/test_create_bank_request.py
index 10bf2052..0e7d8157 100644
--- a/hindsight-clients/python/hindsight_client_api/test/test_create_bank_request.py
+++ b/hindsight-clients/python/hindsight_client_api/test/test_create_bank_request.py
@@ -36,7 +36,7 @@ class TestCreateBankRequest(unittest.TestCase):
         if include_optional:
             return CreateBankRequest(
                 name = '',
-                disposition = {agreeableness=0.7, bias_strength=0.7, conscientiousness=0.6, extraversion=0.5, neuroticism=0.3, openness=0.8},
+                disposition = {empathy=3, literalism=3, skepticism=3},
                 background = ''
             )
         else:
diff --git a/hindsight-clients/python/hindsight_client_api/test/test_disposition_traits.py b/hindsight-clients/python/hindsight_client_api/test/test_disposition_traits.py
index e4ea5a6f..fbc7fd1b 100644
--- a/hindsight-clients/python/hindsight_client_api/test/test_disposition_traits.py
+++ b/hindsight-clients/python/hindsight_client_api/test/test_disposition_traits.py
@@ -35,21 +35,15 @@ class TestDispositionTraits(unittest.TestCase):
         model = DispositionTraits()
         if include_optional:
             return DispositionTraits(
-                openness = 0.0,
-                conscientiousness = 0.0,
-                extraversion = 0.0,
-                agreeableness = 0.0,
-                neuroticism = 0.0,
-                bias_strength = 0.0
+                skepticism = 1.0,
+                literalism = 1.0,
+                empathy = 1.0
             )
         else:
             return DispositionTraits(
-                openness = 0.0,
-                conscientiousness = 0.0,
-                extraversion = 0.0,
-                agreeableness = 0.0,
-                neuroticism = 0.0,
-                bias_strength = 0.0,
+                skepticism = 1.0,
+                literalism = 1.0,
+                empathy = 1.0,
         )
         """
 
diff --git a/hindsight-clients/python/hindsight_client_api/test/test_update_disposition_request.py b/hindsight-clients/python/hindsight_client_api/test/test_update_disposition_request.py
index 1686ce21..228f019d 100644
--- a/hindsight-clients/python/hindsight_client_api/test/test_update_disposition_request.py
+++ b/hindsight-clients/python/hindsight_client_api/test/test_update_disposition_request.py
@@ -35,11 +35,11 @@ class TestUpdateDispositionRequest(unittest.TestCase):
         model = UpdateDispositionRequest()
         if include_optional:
             return UpdateDispositionRequest(
-                disposition = {agreeableness=0.7, bias_strength=0.7, conscientiousness=0.6, extraversion=0.5, neuroticism=0.3, openness=0.8}
+                disposition = {empathy=3, literalism=3, skepticism=3}
             )
         else:
             return UpdateDispositionRequest(
-                disposition = {agreeableness=0.7, bias_strength=0.7, conscientiousness=0.6, extraversion=0.5, neuroticism=0.3, openness=0.8},
+                disposition = {empathy=3, literalism=3, skepticism=3},
         )
         """
 
diff --git a/hindsight-clients/rust/Cargo.lock b/hindsight-clients/rust/Cargo.lock
index 98cc87d1..e158a2b4 100644
--- a/hindsight-clients/rust/Cargo.lock
+++ b/hindsight-clients/rust/Cargo.lock
@@ -86,9 +86,9 @@ checksum = "b35204fbdc0b3f4446b89fc1ac2cf84a8a68971995d0bf2e925ec7cd960f9cb3"
 
 [[package]]
 name = "cc"
-version = "1.2.48"
+version = "1.2.49"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c481bdbf0ed3b892f6f806287d72acd515b352a4ec27a208489b8c1bc839633a"
+checksum = "90583009037521a116abf44494efecd645ba48b6622457080f080b85544e2215"
 dependencies = [
  "find-msvc-tools",
  "shlex",
diff --git a/hindsight-clients/typescript/generated/sdk.gen.ts b/hindsight-clients/typescript/generated/sdk.gen.ts
index c1b1fb2c..80d05893 100644
--- a/hindsight-clients/typescript/generated/sdk.gen.ts
+++ b/hindsight-clients/typescript/generated/sdk.gen.ts
@@ -183,7 +183,7 @@ export const getBankProfile = <ThrowOnError extends boolean = false>(options: Op
 /**
  * Update memory bank disposition
  *
- * Update bank's Big Five disposition traits and bias strength
+ * Update bank's disposition traits (skepticism, literalism, empathy)
  */
 export const updateBankDisposition = <ThrowOnError extends boolean = false>(options: Options<UpdateBankDispositionData, ThrowOnError>) => (options.client ?? client).put<UpdateBankDispositionResponses, UpdateBankDispositionErrors, ThrowOnError>({
     url: '/v1/default/banks/{bank_id}/profile',
diff --git a/hindsight-clients/typescript/generated/types.gen.ts b/hindsight-clients/typescript/generated/types.gen.ts
index 55561b48..2536b9ab 100644
--- a/hindsight-clients/typescript/generated/types.gen.ts
+++ b/hindsight-clients/typescript/generated/types.gen.ts
@@ -19,7 +19,7 @@ export type AddBackgroundRequest = {
     /**
      * Update Disposition
      *
-     * If true, infer Big Five disposition traits from the merged background (default: true)
+     * If true, infer disposition traits from the merged background (default: true)
      */
     update_disposition?: boolean;
 };
@@ -210,45 +210,27 @@ export type DeleteResponse = {
 /**
  * DispositionTraits
  *
- * Disposition traits based on Big Five model.
+ * Disposition traits that influence how memories are formed and interpreted.
  */
 export type DispositionTraits = {
     /**
-     * Openness
+     * Skepticism
      *
-     * Openness to experience (0-1)
+     * How skeptical vs trusting (1=trusting, 5=skeptical)
      */
-    openness: number;
+    skepticism: number;
     /**
-     * Conscientiousness
+     * Literalism
      *
-     * Conscientiousness (0-1)
+     * How literally to interpret information (1=flexible, 5=literal)
      */
-    conscientiousness: number;
+    literalism: number;
     /**
-     * Extraversion
+     * Empathy
      *
-     * Extraversion (0-1)
+     * How much to consider emotional context (1=detached, 5=empathetic)
      */
-    extraversion: number;
-    /**
-     * Agreeableness
-     *
-     * Agreeableness (0-1)
-     */
-    agreeableness: number;
-    /**
-     * Neuroticism
-     *
-     * Neuroticism (0-1)
-     */
-    neuroticism: number;
-    /**
-     * Bias Strength
-     *
-     * How strongly disposition influences opinions (0-1)
-     */
-    bias_strength: number;
+    empathy: number;
 };
 
 /**
diff --git a/hindsight-dev/benchmarks/common/benchmark_runner.py b/hindsight-dev/benchmarks/common/benchmark_runner.py
index 7b0aea9d..121b14d7 100644
--- a/hindsight-dev/benchmarks/common/benchmark_runner.py
+++ b/hindsight-dev/benchmarks/common/benchmark_runner.py
@@ -471,7 +471,9 @@ class BenchmarkRunner:
                 fact_type=["world", "experience"],
                 question_date=question_date,
                 include_entities=True,
-                include_chunks=True
+                max_entity_tokens=2048,
+                include_chunks=True,
+
             )
             recall_time = time.time() - recall_start_time
 
diff --git a/openapi.json b/openapi.json
index f3157f5c..7d0376ab 100644
--- a/openapi.json
+++ b/openapi.json
@@ -844,7 +844,7 @@
       },
       "put": {
         "summary": "Update memory bank disposition",
-        "description": "Update bank's Big Five disposition traits and bias strength",
+        "description": "Update bank's disposition traits (skepticism, literalism, empathy)",
         "operationId": "update_bank_disposition",
         "parameters": [
           {
@@ -1110,7 +1110,7 @@
           "update_disposition": {
             "type": "boolean",
             "title": "Update Disposition",
-            "description": "If true, infer Big Five disposition traits from the merged background (default: true)",
+            "description": "If true, infer disposition traits from the merged background (default: true)",
             "default": true
           }
         },
@@ -1151,12 +1151,9 @@
         "example": {
           "background": "I was born in Texas. I am a software engineer with 10 years of experience.",
           "disposition": {
-            "agreeableness": 0.8,
-            "bias_strength": 0.6,
-            "conscientiousness": 0.6,
-            "extraversion": 0.5,
-            "neuroticism": 0.4,
-            "openness": 0.7
+            "empathy": 3,
+            "literalism": 3,
+            "skepticism": 3
           }
         }
       },
@@ -1233,12 +1230,9 @@
               "bank_id": "user123",
               "created_at": "2024-01-15T10:30:00Z",
               "disposition": {
-                "agreeableness": 0.5,
-                "bias_strength": 0.5,
-                "conscientiousness": 0.5,
-                "extraversion": 0.5,
-                "neuroticism": 0.5,
-                "openness": 0.5
+                "empathy": 3,
+                "literalism": 3,
+                "skepticism": 3
               },
               "name": "Alice",
               "updated_at": "2024-01-16T14:20:00Z"
@@ -1277,12 +1271,9 @@
           "background": "I am a software engineer with 10 years of experience in startups",
           "bank_id": "user123",
           "disposition": {
-            "agreeableness": 0.7,
-            "bias_strength": 0.7,
-            "conscientiousness": 0.6,
-            "extraversion": 0.5,
-            "neuroticism": 0.3,
-            "openness": 0.8
+            "empathy": 3,
+            "literalism": 3,
+            "skepticism": 3
           },
           "name": "Alice"
         }
@@ -1428,12 +1419,9 @@
         "example": {
           "background": "I am a creative software engineer with 10 years of experience",
           "disposition": {
-            "agreeableness": 0.7,
-            "bias_strength": 0.7,
-            "conscientiousness": 0.6,
-            "extraversion": 0.5,
-            "neuroticism": 0.3,
-            "openness": 0.8
+            "empathy": 3,
+            "literalism": 3,
+            "skepticism": 3
           },
           "name": "Alice"
         }
@@ -1457,67 +1445,40 @@
       },
       "DispositionTraits": {
         "properties": {
-          "openness": {
-            "type": "number",
-            "maximum": 1.0,
-            "minimum": 0.0,
-            "title": "Openness",
-            "description": "Openness to experience (0-1)"
+          "skepticism": {
+            "type": "integer",
+            "maximum": 5.0,
+            "minimum": 1.0,
+            "title": "Skepticism",
+            "description": "How skeptical vs trusting (1=trusting, 5=skeptical)"
           },
-          "conscientiousness": {
-            "type": "number",
-            "maximum": 1.0,
-            "minimum": 0.0,
-            "title": "Conscientiousness",
-            "description": "Conscientiousness (0-1)"
+          "literalism": {
+            "type": "integer",
+            "maximum": 5.0,
+            "minimum": 1.0,
+            "title": "Literalism",
+            "description": "How literally to interpret information (1=flexible, 5=literal)"
           },
-          "extraversion": {
-            "type": "number",
-            "maximum": 1.0,
-            "minimum": 0.0,
-            "title": "Extraversion",
-            "description": "Extraversion (0-1)"
-          },
-          "agreeableness": {
-            "type": "number",
-            "maximum": 1.0,
-            "minimum": 0.0,
-            "title": "Agreeableness",
-            "description": "Agreeableness (0-1)"
-          },
-          "neuroticism": {
-            "type": "number",
-            "maximum": 1.0,
-            "minimum": 0.0,
-            "title": "Neuroticism",
-            "description": "Neuroticism (0-1)"
-          },
-          "bias_strength": {
-            "type": "number",
-            "maximum": 1.0,
-            "minimum": 0.0,
-            "title": "Bias Strength",
-            "description": "How strongly disposition influences opinions (0-1)"
+          "empathy": {
+            "type": "integer",
+            "maximum": 5.0,
+            "minimum": 1.0,
+            "title": "Empathy",
+            "description": "How much to consider emotional context (1=detached, 5=empathetic)"
           }
         },
         "type": "object",
         "required": [
-          "openness",
-          "conscientiousness",
-          "extraversion",
-          "agreeableness",
-          "neuroticism",
-          "bias_strength"
+          "skepticism",
+          "literalism",
+          "empathy"
         ],
         "title": "DispositionTraits",
-        "description": "Disposition traits based on Big Five model.",
+        "description": "Disposition traits that influence how memories are formed and interpreted.",
         "example": {
-          "agreeableness": 0.7,
-          "bias_strength": 0.7,
-          "conscientiousness": 0.6,
-          "extraversion": 0.5,
-          "neuroticism": 0.3,
-          "openness": 0.8
+          "empathy": 3,
+          "literalism": 3,
+          "skepticism": 3
         }
       },
       "DocumentResponse": {