Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Entropy

A Quantum Model of Trust Calibration in Human–AI Interactions

Luisa Roeder, Pamela Hoyte, Johan N. van der Meer, Lauren Fell, Patrick Johnston, Graham Kerr, Peter Bruza

Published Sep 20, 2023• 19 citations
Publisher

Abstract

This exploratory study investigates a human agent's evolving judgements of reliability when interacting with an AI system. Two aims drove this investigation: (1) compare the predictive performance of quantum vs. Markov random walk models regarding human reliability judgements of an AI system and (2) identify a neural correlate of the perturbation of a human agent's judgement of the AI's reliability. As AI becomes more prevalent, it is important to understand how humans trust these technologies and how trust evolves when interacting with them. A mixed-methods experiment was developed for exploring reliability calibration in human-AI interactions. The behavioural data collected were used as a baseline to assess the predictive performance of the quantum and Markov models. We found the quantum model to better predict the evolving reliability ratings than the Markov model. This may be due to the quantum model being more amenable to represent the sometimes pronounced within-subject variability of reliability ratings. Additionally, a clear event-related potential response was found in the electroencephalographic (EEG) data, which is attributed to the expectations of reliability being perturbed. The identification of a trust-related EEG-based measure opens the door to explore how it could be used to adapt the parameters of the quantum model in real time.

DOI
10.3390/e25091362

Related papers

Modelling the quantum-like dynamics of human reliability ratings in human–AI interactions by interaction-dependent Hamiltonians

Johan N. van der Meer, Pamela Hoyte, Luisa Roeder +1 more

We are rapidly moving to a future in which our information environment is saturated by artificial intelligence (AI), and humans and AI agents will routinely engage in shared decision making even in conditions of high uncertainty and risk (such as natural disasters or nuclear accidents). Trust is fundamental to the effectiveness of these interactions. A key challenge in modelling the dynamics of trust in human-AI interactions is to provide a means to integrate the diversity of human trust fluctuations found empirically. In this article, we explore the ability of quantum random walk (QRW) models to model this dynamism of trust found in empirical human-AI interactions. Specifically, we manipulate certain features of the QRW to explore its ability to provide the necessary agility and sensitivity to fluctuations in trust judgments. The goal is to incorporate the nature of the interaction itself into the evolution of the model, with the features stemming from empirically derived parameters. We found that using empirical parameters to inform the use of different Hamiltonians throughout the interaction markedly influences the modelled trust dynamics and can provide a promising means to model the evolution of trust in human-AI interactions.This article is part of the theme issue 'Quantum theory and topology in models of decision making (Part 1)'.

arXiv:2504.13918•Philosophical Transactions of the Royal Society A Mathematical Physical and Engineering Sciences•Nov 27, 2025•5 citations

Comparing quantum versus Markov random walk models of judgements measured by rating scales

Zisu Wang, Jerome R. Busemeyer

Quantum and Markov random walk models are proposed for describing how people evaluate stimuli using rating scales. To empirically test these competing models, we conducted an experiment in which participants judged the effectiveness of public health service announcements from either their own personal perspective or from the perspective of another person. The order of the self versus other judgements was manipulated, which produced significant sequential effects. The quantum and Markov models were fitted to the data using the same number of parameters, and the model comparison strongly supported the quantum over the Markov model.

Philosophical Transactions of the Royal Society A Mathematical Physical and Engineering Sciences•Nov 30, 2015•23 citations

Algorithmic Transparency and Explainable AI using Quantum Information Theory

Jincheng Zhang

The increasing deployment of Artificial Intelligence (AI) systems across various domains necessitates a deeper understanding of their decision-making processes. Current explainable AI (XAI) techniques often struggle to provide truly transparent and reliable explanations, particularly for complex models like deep neural networks. This paper proposes a novel approach to algorithmic transparency and XAI leveraging concepts from quantum information theory. Specifically, we explore the potential of quantum entanglement and mutual information to represent and analyze AI model decisions in a fundamentally different way. Our framework represents model outputs as quantum states and investigates the correlations between these states and the input data. This analysis aims to identify the key factors driving the model's decision and provide a more intuitive and understandable explanation. We demonstrate that utilizing quantum information theory can offer a richer and more robust approach to XAI than traditional methods, potentially addressing limitations related to interpretability and trust in AI systems. This work lays the groundwork for a new paradigm in AI transparency, utilizing the unique properties of quantum mechanics to enhance the explainability of complex algorithms.

•Aug 28, 2026•0 citations

Reconciling Irrational Human Behavior with AI based Decision Making: A Quantum Probabilistic Approach

Sagar Uprety, Dawei Song

There are many examples of human decision making which cannot be modeled by classical probabilistic and logic models, on which the current AI systems are based. Hence the need for a modeling framework which can enable intelligent systems to detect and predict cognitive biases in human decisions to facilitate better human-agent interaction. We give a few examples of irrational behavior and use a generalized probabilistic model inspired by the mathematical framework of Quantum Theory to model and explain such behavior.

arXiv:1808.04600•Aug 14, 2018•2 citations

Hybrid Quantum–AI Models for Healthcare Prediction: Enhancing Accuracy on Complex Data and Addressing Barriers to Clinical Adoption

SOHAN LAL GUPTA, Dr. Vipin Jain, Dr. Arpita Sharma +2 more

The rapid growth of healthcare data demands predictive models capable of handling complexity, noise, and imbalance inherent in clinical datasets. This study explores the integration of quantum computing with artificial intelligence to develop Hybrid Quantum–AI models for healthcare prediction. The proposed framework combines classical preprocessing and feature extraction with quantum neural networks (QNNs) and quantum kernel methods to enhance predictive performance on diverse healthcare tasks, including disease diagnosis and postoperative complication prediction. Empirical evaluation across benchmark and clinical datasets demonstrates that hybrid quantum models achieve superior sensitivity, precision, and calibration compared to traditional machine learning approaches, particularly under conditions of data imperfection and small sample size. Beyond performance, this work investigates practical barriers to clinical adoption, including hardware limitations, scalability, interpretability, and ethical compliance. The results highlight that while current quantum hardware remains a constraint, hybrid approaches already offer tangible benefits in predictive accuracy and robustness. The study concludes by outlining a roadmap for real-world implementation, emphasizing the need for interpretable hybrid architectures, federated data strategies, and regulatory alignment to enable the transition of quantum–AI healthcare solutions from research to clinical practice.

•Dec 5, 2025•1 citation

Application of Quantum—Markov Open System Models to Human Cognition and Decision

Jerome R. Busemeyer, Qizi Zhang, Sivasubramanya Nadar Balakrishnan +1 more

Markov processes, such as random walk models, have been successfully used by cognitive and neural scientists to model human choice behavior and decision time for over 50 years. Recently, quantum walk models have been introduced as an alternative way to model the dynamics of human choice and confidence across time. Empirical evidence points to the need for both types of processes, and open system models provide a way to incorporate them both into a single process. However, some of the constraints required by open system models present challenges for achieving this goal. The purpose of this article is to address these challenges and formulate open system models that have good potential to make important advancements in cognitive science.

Entropy•Sep 4, 2020•27 citations
This exploratory study investigates a human agent's evolving judgements of reliability when interacting with an AI system. Two aims drove this investigation: (1) compare the predictive performance of quantum vs. Markov random walk models regarding human reliability judgements of an AI system and (2) identify a neural correlate of the perturbation of a human agent's judgement of the AI's reliability. As AI becomes more prevalent, it is important to understand how humans trust these technologies and how trust evolves when interacting with them. A mixed-methods experiment was developed for exploring reliability calibration in human-AI interactions. The behavioural data collected were used as a baseline to assess the predictive performance of the quantum and Markov models. We found the quantum model to better predict the evolving reliability ratings than the Markov model. This may be due to the quantum model being more amenable to represent the sometimes pronounced within-subject variability of reliability ratings. Additionally, a clear event-related potential response was found in the electroencephalographic (EEG) data, which is attributed to the expectations of reliability being perturbed. The identification of a trust-related EEG-based measure opens the door to explore how it could be used to adapt the parameters of the quantum model in real time.
We are rapidly moving to a future in which our information environment is saturated by artificial intelligence (AI), and humans and AI agents will routinely engage in shared decision making even in conditions of high uncertainty and risk (such as natural disasters or nuclear accidents). Trust is fundamental to the effectiveness of these interactions. A key challenge in modelling the dynamics of trust in human-AI interactions is to provide a means to integrate the diversity of human trust fluctuations found empirically. In this article, we explore the ability of quantum random walk (QRW) models to model this dynamism of trust found in empirical human-AI interactions. Specifically, we manipulate certain features of the QRW to explore its ability to provide the necessary agility and sensitivity to fluctuations in trust judgments. The goal is to incorporate the nature of the interaction itself into the evolution of the model, with the features stemming from empirically derived parameters. We found that using empirical parameters to inform the use of different Hamiltonians throughout the interaction markedly influences the modelled trust dynamics and can provide a promising means to model the evolution of trust in human-AI interactions.This article is part of the theme issue 'Quantum theory and topology in models of decision making (Part 1)'.
Quantum and Markov random walk models are proposed for describing how people evaluate stimuli using rating scales. To empirically test these competing models, we conducted an experiment in which participants judged the effectiveness of public health service announcements from either their own personal perspective or from the perspective of another person. The order of the self versus other judgements was manipulated, which produced significant sequential effects. The quantum and Markov models were fitted to the data using the same number of parameters, and the model comparison strongly supported the quantum over the Markov model.
The increasing deployment of Artificial Intelligence (AI) systems across various domains necessitates a deeper understanding of their decision-making processes. Current explainable AI (XAI) techniques often struggle to provide truly transparent and reliable explanations, particularly for complex models like deep neural networks. This paper proposes a novel approach to algorithmic transparency and XAI leveraging concepts from quantum information theory. Specifically, we explore the potential of quantum entanglement and mutual information to represent and analyze AI model decisions in a fundamentally different way. Our framework represents model outputs as quantum states and investigates the correlations between these states and the input data. This analysis aims to identify the key factors driving the model's decision and provide a more intuitive and understandable explanation. We demonstrate that utilizing quantum information theory can offer a richer and more robust approach to XAI than traditional methods, potentially addressing limitations related to interpretability and trust in AI systems. This work lays the groundwork for a new paradigm in AI transparency, utilizing the unique properties of quantum mechanics to enhance the explainability of complex algorithms.
There are many examples of human decision making which cannot be modeled by classical probabilistic and logic models, on which the current AI systems are based. Hence the need for a modeling framework which can enable intelligent systems to detect and predict cognitive biases in human decisions to facilitate better human-agent interaction. We give a few examples of irrational behavior and use a generalized probabilistic model inspired by the mathematical framework of Quantum Theory to model and explain such behavior.
The rapid growth of healthcare data demands predictive models capable of handling complexity, noise, and imbalance inherent in clinical datasets. This study explores the integration of quantum computing with artificial intelligence to develop Hybrid Quantum–AI models for healthcare prediction. The proposed framework combines classical preprocessing and feature extraction with quantum neural networks (QNNs) and quantum kernel methods to enhance predictive performance on diverse healthcare tasks, including disease diagnosis and postoperative complication prediction. Empirical evaluation across benchmark and clinical datasets demonstrates that hybrid quantum models achieve superior sensitivity, precision, and calibration compared to traditional machine learning approaches, particularly under conditions of data imperfection and small sample size. Beyond performance, this work investigates practical barriers to clinical adoption, including hardware limitations, scalability, interpretability, and ethical compliance. The results highlight that while current quantum hardware remains a constraint, hybrid approaches already offer tangible benefits in predictive accuracy and robustness. The study concludes by outlining a roadmap for real-world implementation, emphasizing the need for interpretable hybrid architectures, federated data strategies, and regulatory alignment to enable the transition of quantum–AI healthcare solutions from research to clinical practice.
Markov processes, such as random walk models, have been successfully used by cognitive and neural scientists to model human choice behavior and decision time for over 50 years. Recently, quantum walk models have been introduced as an alternative way to model the dynamics of human choice and confidence across time. Empirical evidence points to the need for both types of processes, and open system models provide a way to incorporate them both into a single process. However, some of the constraints required by open system models present challenges for achieving this goal. The purpose of this article is to address these challenges and formulate open system models that have good potential to make important advancements in cognitive science.