The question of under what conditions conceptual representation is compositional remains debatable within cognitive science.This paper proposes a well developed mathematical apparatus for a probabilistic representation of concepts, drawing upon methods developed in quantum theory to propose a formal test that can determine whether a specific conceptual combination is compositional, or not.This test examines a joint probability distribution modeling the combination, asking whether or not it is factorizable.Empirical studies indicate that some combinations should be considered non-compositionally.
The question of under what conditions conceptual representation is compositional remains debatable within cognitive science.This paper proposes a well developed mathematical apparatus for a probabilistic representation of concepts, drawing upon methods developed in quantum theory to propose a formal test that can determine whether a specific conceptual combination is compositional, or not.This test examines a joint probability distribution modeling the combination, asking whether or not it is factorizable.Empirical studies indicate that some combinations should be considered non-compositionally.
Consider the concept combination “pet human.” In word association experiments, human subjects often produce the associate “slave” in relation to this combination. The striking aspect of this associate is that it is not produced as an associate of “pet” or “human” in isolation. In other words, the associate “slave” cannot be recovered from the constituent concepts. Such examples have been used in both cognitive science and philosophy to argue that concept combinations have a non-compositional semantics. This chapter will feature how various non-compositional accounts of concept combinations can be provided from quantum theory. Quantum theory is a theory which caters for the modelling of non-compositionality because the state of a quantum entangled system cannot be constructed from the states of its individual subsystems. Utilizing probabilistic methods developed for analyzing composite systems in quantum theory, we show that it is possible to classify concept combinations as having “classically compositional,” “pseudo-classically non-compositional,” or “non-classically non-compositional” semantics by determining whether the joint probability distribution modelling the combination is factorizable or not. Concept combinations and cognition The principle of semantic compositionality states the meaning of a (syntactically complex) whole is a function only of the meanings of its (syntactic) parts together with the manner in which these parts were combined (Pelletier, 1994). Whether the semantics of concepts are compositional, or not, has been somewhat of a battleground. On the one hand, there are authors like Fodor (1994) who are adamant that concepts are, and must be, compositional, but this position stands in contrast with the computational linguist Zadrozny (1994), who produced a theorem suggesting that: “the standard definition of compositionality is formally vacuous.”
The representation and processing of concepts is one of the hardest challenges in cognitive science. While computer scientists and engineers have focused on developing advances for particular tasks, philosophers and cognitive scientists have focused on elucidating the structural nature of meaning. A remarkable bridge between these two limited-success approaches can be found in behavioral research, since, in a variety of tasks, humans process information at a conceptual level in a way that is incompatible with classical probability and fuzzy set theory. Recently, attempts have been made to use mathematical schemes founded on quantum structures as alternative approaches. For this reason, the application of quantum structures to this type of phenomena has received increasing attention. The quantum approach allows to faithfully model a number of non-classical deviations observed in experimental data. Moreover, it shows that genuine quantum theoretical notions, such as contextuality, superposition, emergence, and entanglement, are powerful epistemic tools to understand and represent cognitive phenomena. In this thesis, we identify the limitations of classical theories to handle some important cognitive tasks, and introduce the fundamentals of the quantum cognitive approach to concepts. Next, we perform a mathematical analysis of current concept combination models and develop an extension that allows for concrete representations of multiple exemplars simultaneously. Our analysis indicates that a superposition of logical reasoning and a specific form of non-logical reasoning, where non-logical reasoning is dominant, allows to faithfully represent the experimental data. Therefore, the non-logical reasoning introduced by this model represents an important but unexplored form of reasoning in humans. In addition, we develop novel experimental methodologies to identify quantum conceptual structures for concept combinations in the context of natural language processing and psychological experiments. Namely, we present a methodology to build entangled concepts represented as sets of words with respect to a corpus of text, and present a computational and psychological methodology to discern if a collection of concepts behaves statistically as a collection of quantum or classical particles. Using both methodologies we have identified a significant presence of quantum conceptual structure in the context of natural language processing and psychological experiments.
This article introduces a “pseudo classical” notion of modelling non-separability. This form of non-separability can be viewed as lying between separability and quantum-like non-separability. Non-separability is formalized in terms of the non-factorizabilty of the underlying joint probability distribution. A decision criterium for determining the non-factorizability of the joint distribution is related to determining the rank of a matrix as well as another approach based on the chi-square-goodness-of-fit test. This pseudo-classical notion of non-separability is discussed in terms of quantum games and concept combinations in human cognition.
Objectives: This research is first of its kind to investigate the application of quantum mechanics and related mathematical models to cognitive science, specifically focusing on the challenge of modeling how the human mind assigns meaning to combinations of words. The primary objective is to classify whether given word combinations are compositional or non-compositional. Methods: The author introduces the "WWW method," which leverages the World Wide Web (WWW) as an experimental tool in place of human subjects, which makes this method cheaper and handy in comparison to the earlier methods which used human subjects for spoken and sign language studies and robots as non-human subjects. The "WWW method" involves checking the adherence to Bell inequalities and marginal selectivity. The term "meaning bound" is introduced to quantify the extent to which two words influence each other's meaning within a combination. This concept proves instrumental in devising a method for assessing compositionality using the WWW method. Findings: When the WWW serves as the information space, the verification of whether a word combination adheres to Bell inequalities entails determining if the meaning bound for the word combination (AB & C) exceeds that of its constituent parts (A & C, B & C) within the conceptual combination AB. Marginal selectivity is evaluated by establishing a vector model for grammatically framed word combinations, utilizing the WWW to collect data. If the word combination 'A then B' yields an equivalent number of hits as 'B then A,' it signifies that the word combination AB satisfies marginal selectivity. Novelty: Consequently, this research presents an accessible approach to identifying entangled pairs of word combinations using the WWW method, eliminating the need for human subject experiments. The earlier studies which include human subjects are costly and cannot be replicated anytime-anywhere, for obvious reasons. Keywords: Quantum cognition, Marginal selectivity, Compositionality, WWW method, CHSH inequality
The authors provide an overview of the results they have obtained over the last decade on the identification of quantum structures in cognition and, more specifically, in the formalization and representation of natural concepts. They firstly discuss the quantum foundational reasons that led them to investigate the mechanisms of formation and combination of concepts in human reasoning, starting from the empirically observed deviations from classical, logical, and probabilistic structures. They then develop their quantum-theoretic perspective in Fock space which allows successful modeling of various sets of cognitive experiments collected by different scientists, including themselves. In addition, they formulate a unified explanatory hypothesis for the presence of quantum structures in cognitive processes, and discuss their recent discovery of further quantum aspects in concept combinations, namely, ‘entanglement’ and ‘indistinguishability.’ They finally illustrate the outlook for future research.
Experiments in cognitive science and decision theory show that the ways in which people combine concepts and make decisions cannot be described by classical logic and probability theory. This has serious implications for applied disciplines such as information retrieval, artificial intelligence and robotics. Inspired by a mathematical formalism that generalizes quantum mechanics the authors have constructed a contextual framework for both concept representation and decision making, together with quantum models that are in strong alignment with experimental data. The results can be interpreted by assuming the existence in human thought of a double-layered structure, a 'classical logical thought' and a 'quantum conceptual thought', the latter being responsible of the above paradoxes and nonclassical effects. The presence of a quantum structure in cognition is relevant, for it shows that quantum mechanics provides not only a useful modeling tool for experimental data but also supplies a structural model for human and artificial thought processes. This approach has strong connections with theories formalizing meaning, such as semantic analysis, and has also a deep impact on computer science, information retrieval and artificial intelligence. More specifically, the links with information retrieval are discussed in this paper.