Documentation Index
Fetch the complete documentation index at: https://docs.strategist.gg/llms.txt
Use this file to discover all available pages before exploring further.
Behavioral Psychology Integration
Nash Equilibrium theory provides the mathematical foundation, but human psychology determines practical implementation. THE STRATEGIST integrates contemporary behavioral research to bridge the gap between theoretical optimality and real-world game-theoretic implementation.
Signaling Games for Personal Development
Spence’s Job Market Model Applied to Personal Strategy
From Spence’s Nobel Prize-winning work: Workers signal productivity through education costs that high-ability workers can bear more easily than low-ability workers.
Personal Signaling Framework:
Signal: s ∈ [0, ∞) (intensity of strategic behavior)
Type: θ ∈ {High Discipline, Low Discipline}
Cost: c(s, θ) where c(s, High) < c(s, Low) for s > 0
Separating Equilibrium: Different types choose different effort levels
Pooling Equilibrium: All types choose similar strategies
Signaling in Personal Contexts
Exercise Routine Signaling:
- Signal: Intensity and consistency of workout routine
- High Type: Intrinsically motivated individuals
- Low Type: Externally motivated individuals
- Separating Equilibrium: High types choose challenging routines (CrossFit, marathons), low types choose moderate routines (walking, yoga)
- Strategic Value: Reveals true motivation level to yourself and others
Professional Signaling:
- Signal: Side projects, continuous learning, early arrival
- High Type: Naturally productive individuals
- Low Type: Less naturally productive individuals
- Separating Equilibrium: High types invest in visible competence signals, low types focus on efficient minimum viable performance
Mathematical Signaling Model
Utility Functions:
High Type: U_H(s, w) = w - c_H(s)
Low Type: U_L(s, w) = w - c_L(s)
Where:
w = wage/benefit from signaling level s
c_H(s) < c_L(s) = cost of signaling (lower for high type)
Incentive Compatibility (High Type):
U_H(s_H, w_H) ≥ U_H(s_L, w_L)
Incentive Compatibility (Low Type):
U_L(s_L, w_L) ≥ U_L(s_H, w_H)
Individual Rationality:
U_H(s_H, w_H) ≥ 0 and U_L(s_L, w_L) ≥ 0
Bounded Rationality Models
Herbert Simon’s Satisficing Behavior
Instead of maximizing utility, humans seek “good enough” solutions that exceed an aspiration level.
Satisficing Algorithm:
def satisficing_decision(options, aspiration_level, search_cost):
"""
Search options until finding one that exceeds aspiration level
vs. examining all options to find global maximum
"""
total_search_cost = 0
for option in random.shuffle(options):
total_search_cost += search_cost
if evaluate_option(option) >= aspiration_level:
return option, total_search_cost
# If no satisficing option found, lower aspiration level
return satisficing_decision(options, aspiration_level * 0.9, search_cost)
- Morning Routine: Find “good enough” routine rather than perfect optimization
- Career Choices: Satisficing job selection vs. exhaustive evaluation
- Social Plans: Accept first acceptable social option vs. optimal social configuration
Cognitive Constraints in Strategic Thinking
Working Memory Limits: 7±2 items in strategic calculations
Level-k Thinking: Users typically think 1-2 steps ahead
- Level 0: Random or instinctive actions
- Level 1: Best response to Level 0 thinking
- Level 2: Best response to Level 1 thinking
- Strategic Implication: Most people are Level 1-2 thinkers; assuming higher-level thinking leads to strategic failures
Present Bias: Exponential discounting with present preference
Utility = u(today) + β * δ * u(tomorrow) + β * δ² * u(day_after)
Where:
β < 1 = present bias parameter (additional discount on future)
δ = standard discount factor
Multi-Agent Psychological Realism
System 1 vs System 2 Integration
From Kahneman’s dual-process theory, mapped to internal agent conflicts:
Fast Agents (System 1):
- Protector: Immediate safety responses, threat detection
- Connector: Automatic social harmony, relationship maintenance
Slow Agents (System 2):
- Optimizer: Analytical efficiency calculation, deliberate optimization
- Explorer: Long-term planning, strategic growth assessment
Agent Activation Patterns
Stress Response Hierarchy:
def stress_response_agent_activation(stress_level):
"""
Higher stress shifts control to System 1 agents
Lower stress enables System 2 agent coordination
"""
if stress_level > 8: # Crisis mode
return {
'protector': 0.70, # Dominant safety focus
'connector': 0.20, # Seek social support
'optimizer': 0.06, # Minimal analytical capacity
'explorer': 0.04 # Almost no growth orientation
}
elif stress_level > 5: # Moderate stress
return {
'protector': 0.40,
'connector': 0.25,
'optimizer': 0.25,
'explorer': 0.10
}
else: # Low stress - optimal for strategic thinking
return {
'protector': 0.20,
'connector': 0.25,
'optimizer': 0.30,
'explorer': 0.25
}
Emotional State Game Theory
Emotions as Strategic Information:
- Anger: Signals commitment to punish defection
- Guilt: Signals commitment to cooperation
- Fear: Signals vulnerability requiring protection
- Joy: Signals successful strategy worth repeating
Emotional Equilibrium:
Emotional Strategy Profile: e* = (e₁*, e₂*, e₃*, e₄*)
Where eᵢ* = optimal emotional expression for agent i
Given others' emotional strategies e*₋ᵢ
Commitment Devices and Self-Control
Ulysses Contracts for Strategic Implementation
Mathematical Model:
Commitment Value = E[U(strategic_action)] - Cost_of_commitment - Temptation_resistance_value
1. Financial Commitment
class FinancialCommitment:
def __init__(self, goal, stake_amount, success_criteria):
self.goal = goal
self.stake = stake_amount
self.criteria = success_criteria
def evaluate_commitment(self, actual_behavior):
if self.criteria.is_met(actual_behavior):
return self.stake # Return stake
else:
return -self.stake # Lose stake
2. Social Commitment
class SocialCommitment:
def __init__(self, goal, social_network, reputation_value):
self.goal = goal
self.network = social_network
self.reputation_cost = reputation_value
def social_pressure_utility(self, action):
if action.aligns_with(self.goal):
return self.reputation_cost * 0.1 # Small reputation boost
else:
return -self.reputation_cost # Large reputation loss
3. Temporal Commitment
class TemporalCommitment:
def __init__(self, cooling_off_period):
self.cooling_off = cooling_off_period
def evaluate_action(self, desired_action, current_time):
if current_time < self.cooling_off:
return "BLOCKED" # Prevent impulsive action
else:
return desired_action
Prospect Theory and Strategic Framing
Kahneman-Tversky Value Function Applied to Strategy
Key Behavioral Insights:
- Loss Aversion: Losses feel 2x worse than equivalent gains
- Reference Point Dependence: Outcomes evaluated relative to reference point
- Probability Weighting: Small probabilities overweighted, large probabilities underweighted
Strategic Framing Examples:
Loss vs. Gain Framing
def frame_strategic_choice(current_state, potential_outcome):
"""
Frame strategic choices to align with human psychology
"""
# Loss frame (more motivating for action)
loss_frame = f"Without action, you'll lose {current_state - potential_outcome} strategic advantage"
# Gain frame (less motivating)
gain_frame = f"With action, you'll gain {potential_outcome - current_state} strategic advantage"
# Use loss frame for high-impact strategic decisions
if abs(current_state - potential_outcome) > significance_threshold:
return loss_frame
else:
return gain_frame
Reference Point Anchoring
def set_strategic_reference_points(current_performance, peer_performance, ideal_performance):
"""
Strategic reference point selection affects motivation and satisfaction
"""
reference_options = {
'past_self': current_performance,
'peer_comparison': peer_performance,
'ideal_self': ideal_performance
}
# Choose reference point based on strategic goals
if goal == 'motivation':
return reference_options['ideal_self'] # Creates aspiration
elif goal == 'satisfaction':
return reference_options['past_self'] # Shows progress
elif goal == 'competitive':
return reference_options['peer_comparison'] # Social comparison
Hyperbolic Discounting and Time Preferences
β-δ Model of Present Bias
Standard Exponential: U = u₀ + δu₁ + δ²u₂ + δ³u₃ + …
Hyperbolic with Present Bias: U = u₀ + βδu₁ + βδ²u₂ + βδ³u₃ + …
Where β < 1 creates additional discounting of all future periods.
Strategic Implications
Time-Inconsistent Preferences: Today’s strategic plan differs from tomorrow’s strategic plan
Example:
def strategic_plan_consistency_check(current_plan, future_plan, beta, delta):
"""
Check if strategic plans remain consistent over time
"""
# Current preference for tomorrow vs. day after tomorrow
current_preference = beta * delta * utility_tomorrow + beta * delta**2 * utility_day_after
# Tomorrow's preference for tomorrow vs. day after tomorrow
future_preference = utility_tomorrow + beta * delta * utility_day_after
consistency_ratio = current_preference / future_preference
if abs(consistency_ratio - 1) > 0.1: # 10% inconsistency threshold
return "TIME_INCONSISTENT_PLAN"
else:
return "CONSISTENT_PLAN"
Commitment Strategies for Time Inconsistency
Sophisticated Agents: Recognize own time inconsistency and plan accordingly
Naive Agents: Don’t recognize own time inconsistency
Strategic Response for Sophisticated Agents:
def sophisticated_commitment_strategy(beta, delta, temptation_utility):
"""
Optimal commitment when you know you'll face temptation
"""
# Calculate future self's temptation
future_temptation = beta * temptation_utility
# Set commitment level to exactly offset future temptation
optimal_commitment = -future_temptation + small_buffer
return optimal_commitment
Social Proof and Network Effects
Asch Conformity Experiments show 37% conformity to obviously wrong group answers.
Strategic Conformity Model:
Utility from action a = intrinsic_utility(a) + social_utility(a, group_action)
Where:
social_utility(a, group_action) = conformity_weight * similarity(a, group_action)
Network Strategic Influence
Graphical Games with Social Network:
class SocialNetworkGame:
def __init__(self, network_adjacency_matrix, conformity_weights):
self.network = network_adjacency_matrix
self.weights = conformity_weights
def utility_with_network_effects(self, player, action, all_actions):
intrinsic_utility = self.base_utility(player, action)
social_utility = 0
for neighbor in self.network[player]:
similarity = self.action_similarity(action, all_actions[neighbor])
social_utility += self.weights[player, neighbor] * similarity
return intrinsic_utility + social_utility
- Observe others’ actions before deciding
- Actions reveal information about private signals
- Can lead to suboptimal herding behavior
Strategic Application:
People copy others’ life strategies without considering whether those strategies fit their unique circumstances.
Behavioral realism ensures strategic recommendations align with human psychology rather than purely rational mathematical models.
Next: Implementation Architecture →