Coupling quantum-like cognition with the neuronal networks within generalized probability theory
Andrei Khrennikov,Masanao Ozawa,Felix Benninger,Oded Shor
2024-10-29
Abstract:The recent years are characterized by intensive applications of the methodology and mathematical apparatus of quantum theory, quantum-like modeling, in cognition, psychology, and decision making. In spite of the successful applications of this approach to a variety of psychological effects, e.g., the order, conjunction, disjunction, and response replicability effects, one may (but need not) feel dissatisfaction due to the absence of clear coupling to the neurophysiological processes in the brain. For the moment, this is just a phenomenological approach. In this paper we construct the quantum-like representation of the networks of communicating neurons. It is based not on standard quantum theory, but on generalized probability theory (GPT) with the emphasis of the operational measurement approach. We employ GPT's version which is based on ordered linear state space (instead of complex Hilbert space). A network of communicating neurons is described as a weighted ordered graph that in turn is encoded by its weight matrix. The state space of weight matrices is embedded in GPT with effect-observables and state updates within measurement instruments theory. The latter plays the crucial role. This GPT based model shows the basic quantum-like effects, as e.g. the order, non-repeatability, and disjunction effects; the latter is also known as interference of decisions. This GPT coupling also supports quantum-like modeling in medical diagnostic for neurological diseases, as depression and epilepsy. Although the paper is concentrated on cognition and neuronal networks, the formalism and methodology can be straightforwardly applied to a variety of biological and social networks.
Physics and Society,Artificial Intelligence,Quantum Physics