This project explores whether quantum‑inspired mathematical models can provide a more suitable framework for describing subtle perceptual and physiological phenomena in human–human interactions. These phenomena include emotional resonance, intuitive clinical judgments, somatic markers, and dyadic synchronization processes—patterns that are difficult to capture with classical psychological or physiological models. Modern HCI sensor technologies (HRV, EDA, respiration, IMU‑based micro‑movements, and prosodic voice analysis) enable continuous, multimodal, and unobtrusive measurement of these processes. The project combines these sensor modalities with classical and quantum‑inspired modeling approaches (quantum probability, Hilbert‑space models, non‑commutative decision structures) to examine whether non‑classical patterns emerge in emotional and intuitive perception. A small proof‑of‑concept mini‑pilot study is planned to assess feasibility, data quality, and the potential of multimodal measurements for capturing resonance and somatic markers in natural interactions. The project aims to build an interdisciplinary bridge between medicine, psychology, HCI, and theoretical modeling, and to provide a foundation for future empirical studies and potential collaborations.
This project explores whether quantum‑inspired mathematical models can provide a more suitable framework for describing subtle perceptual and physiological phenomena in human–human interactions. These phenomena include emotional resonance, intuitive clinical judgments, somatic markers, and dyadic synchronization processes—patterns that are difficult to capture with classical psychological or physiological models. Modern HCI sensor technologies (HRV, EDA, respiration, IMU‑based micro‑movements, and prosodic voice analysis) enable continuous, multimodal, and unobtrusive measurement of these processes. The project combines these sensor modalities with classical and quantum‑inspired modeling approaches (quantum probability, Hilbert‑space models, non‑commutative decision structures) to examine whether non‑classical patterns emerge in emotional and intuitive perception. A small proof‑of‑concept mini‑pilot study is planned to assess feasibility, data quality, and the potential of multimodal measurements for capturing resonance and somatic markers in natural interactions. The project aims to build an interdisciplinary bridge between medicine, psychology, HCI, and theoretical modeling, and to provide a foundation for future empirical studies and potential collaborations.
A long-standing goal in the field of Artificial Intelligence (AI) is to develop systems that can perceive and understand human multimodal language. This requires both the consideration of context in the form of surrounding utterances in a conversation, i.e., context modelling, as well as the impact of different modalities (e.g., linguistic, visual acoustic), i.e., multimodal fusion. In the last few years, significant strides have been made towards the interpretation of human language due to simultaneous advancement in deep learning, data gathering and computing infrastructure. AI models have been investigated to either model interactions across distinct modalities, i.e., linguistic, visual and acoustic, or model interactions across parties in a conversation, achieving unprecedented levels of performance. However, AI models are often designed with only performance as their design target, leaving aside other essential factors such as transparency, interpretability, and how humans understand and reason about cognitive states. In line with this observation, in this dissertation, we develop quantum probabilistic neural models and techniques that allow us to capture rational and irrational cognitive biases, without requiring a priori understanding and identification of them. First, we present a comprehensive empirical comparison of state-of-the-art (SOTA) modality fusion strategies for video sentiment analysis. The findings provide us helpful insights into the development of more effective modality fusion models incorporating quantum-inspired components. Second, we introduce an end-to-end complex-valued neural model for video sentiment analysis, simulating quantum procedural steps, outside of physics, into the neural network modelling paradigm. Third, we investigate non-classical correlations across different modalities. In particular, we describe a methodology to model interactions between image and text for an information retrieval scenario. The results provide us with theoretical and empirical insights to develop a transparent end-to-end probabilistic neural model for video emotion detection in conversations, capturing non-classical correlations across distinct modalities. Fourth, we introduce a theoretical framework to model user's cognitive states underlying their multimodal decision perspectives, and propose a methodology to capture interference of modalities in decision making. Overall, we show that our models advance the SOTA on various affective analysis tasks, achieve high transparency due to the mapping to quantum physics meanings, and improve post-hoc interpretability, unearthing useful and explainable knowledge about cross-modal interactions.
Latent variables play a crucial role in psychometric research, yet traditional models often struggle to address context-dependent effects, ambivalent states, and non-commutative measurement processes. This study proposes a quantum-inspired framework for latent variable modeling that employs Hilbert space representations, allowing questionnaire items to be treated as pure or mixed quantum states. By integrating concepts such as superposition, interference, and non-commutative probabilities, the framework captures cognitive and behavioral phenomena that extend beyond the capabilities of classical methods. To illustrate its potential, we introduce quantum-specific metrics—fidelity, overlap, and von Neumann entropy—as complements to correlation-based measures. We also outline a machine-learning pipeline using complex and real-valued neural networks to handle amplitude and phase information. Results highlight the capacity of quantum-inspired models to reveal order effects, ambivalent responses, and multimodal distributions that remain elusive in standard psychometric approaches. This framework broadens the multivariate analysis theoretical and methodological toolkit, offering a dynamic and context-sensitive perspective on latent constructs while inviting further empirical validation in diverse research settings.
Understanding and predicting human behavior remains a complex challenge across scientific disciplines. This study explores a quantum-inspired approach to behavioral modeling, drawing on analogies from quantum mechanics to identify potential underlying patterns in human actions. Key parallels include the spatial distribution of individuals analogous to electron orbitals, the dual nature of behavior at individual and group levels resembling wave-particle duality, and the aggregation of collective decisions reflecting quantum wave function collapse. By integrating concepts from quantum cognition and quantum decision theory, this work examines the theoretical foundations, modeling techniques, and limitations of applying quantum analogies to behavioral prediction. The findings suggest that quantum-inspired models may
The emergence of quantum computing has prompted researchers to examine potential parallels between quantum mechanical principles and human cognitive processes. This article explores the theoretical framework of quantum cognition, examining how concepts from quantum mechanics, including superposition, entanglement, and measurement, might inform our understanding of human decision-making, memory, and consciousness. While quantum computing operates on fundamentally different physical principles than biological neural networks, the mathematical formalisms of quantum mechanics offer novel perspectives on cognitive phenomena that classical probability theory struggles to explain. This review critically examines the current state of quantum cognition research, its applications in computational modeling, and the ongoing debates regarding the validity of quantum mechanical analogies in psychological theory. The analysis reveals both promising theoretical insights and significant limitations in applying quantum principles to human cognition, highlighting the need for empirical validation and careful distinction between mathematical analogy and physical reality.
Emotional cognition is nonlinear, probabilistic, and interdependent, making it hard to recreate human emotional intelligence in AI systems. The study offers the Quantum-Emotion Neural Network (QENN). This unique Quantum Physics-Inspired AI approach uses aspects of quantum mechanics such as superposition, entanglement, and quantum probability amplitudes in neural network design. Where regular models handle emotions as non-interacting or unchanging parts, QENN represents them using amplitude encoding and quantum-inspired neurons to allow all emotional states to be managed together. Training and evaluation of the model took place on the IndiComics Emotion (ICE) dataset, which has 17,340 multimodal instances over eight emotion classes-happy, Sad, Angry, Fearful, Disgust, Surprised, Neutral, and Mixed emotions. The tests showed QENN achieved better results than CNN and LSTM, with an accuracy of 93.8%, 93.5% precision, 92.9% recall, an F1-score of 93.2%, and an ROC-AUC of 95.8% overall. In contrast, traditional CNN and LSTM models had 84.7% and 85.3% accuracies, and their ROC-AUC was 87.2% and 88.5%, respectively. Evaluation with the confusion matrix showed that precision remained above 90% in all emotion types, even for complex ones like ‘Mixed.’ Also, the training-validation accuracy curves revealed excellent generalisation and limited overfitting, with a learning rate of 0.001 giving the most minor validation loss, equal to 0.21. These outcomes show that quantum-inspired solutions are robust and easy to understand for emotion modelling. This work proves that QENN is a critical step forward in making machines that understand emotions as people do.
Emotion recognition using signals with diverse modalities has remained a prominent area of interest in human-computer interaction, medical applications, and immersive interfaces. Although deep learning has made considerable progress in emotion recognition, consumer-level systems have remained ineffective in measuring the uncertainty of the prediction result, incorporating diverse data types, and distinguishing subtle emotions accurately. In light of these challenges, a revolutionary approach using Amplitude, Graded Quantum State Embeddings (AGQSE), and Entangled Correlation Learning (ECL) is proposed. AGQSE maps multimodal inputs such as EEG signals, facial expressions, and speech features into quantum, quantum state embeddings that are normalized, with the level of amplitudes varying based on the credibility and intensity of each component. Individually, these solutions contribute to an advanced mechanism of classifying emotions in a hierarchical manner that yields greater resistance to noisy environments and situations where data is incomplete. Benchmark dataset evaluations corroborate the effectiveness of the proposed method, as evidenced by a leap in precision, intermodal harmony, and uncertainty management against traditional fusion-based solutions. Our framework serves as a stepping stone for the infusion of ideas inspired by quantum mechanics into the domain of affective computing, thereby opening up possibilities in areas such as psychological wellness, smart education, and futuristic human-machine interactions.