A Quantum-Like Model for Predicting Human Decisions in the Entangled Social Systems

Area of Science:

• Decision theory
• Quantum information science
• Artificial intelligence

Background:

• Human-centered systems (e.g., social networks, IoT, healthcare) are increasingly vital.
• Accurate modeling of human behavior under uncertainty is crucial but challenging for conventional probabilistic models.
• Quantum-like decision theories show promise in explaining behavioral anomalies.

Purpose of the Study:

• To develop a predictive quantum-like decision theory for human behavior.
• To integrate quantum entanglement into Bayesian networks (BNs) for enhanced modeling.
• To capture the societal influence on decision-making dynamics.

Main Methods:

• Introduced entanglement from quantum information theory into Bayesian networks, creating an entangled quantum-like BN (QBN).
• Modeled societal effects on decision-making using entanglement measures.
• Developed a quantum-like witness and related it to the concurrence entanglement measure.
• Proposed a predictive entangled QBN (PEQBN) and evaluated it on 22 experimental tasks.

Main Results:

• The PEQBN demonstrated more realistic predictions of human decisions under uncertainty.
• Performance was superior compared to classical BNs and three other quantum-like approaches.
• Entanglement measures effectively modeled societal influences on decision-making.

Conclusions:

• The PEQBN framework provides a novel and effective method for predicting human behavior in complex systems.
• Integrating quantum concepts like entanglement enhances the predictive power of decision models.
• This approach offers a significant advancement in understanding and modeling human decision-making under uncertainty.