Creative Thinking, Familiarity, and Quantum Models

This post is part 1 of a series. It is coauthored by Liane Gabora, Emmanuel Pothos, and Iana Bashmakova.

There is a trend toward increased openness in science, and one aspect of this involves sharing plans and expectations for research projects before they are carried out. (That way, it’s not possible to reframe the results to make it look like they were expected even if they were not.) In this spirit, this post is about research that hasn’t quite happened yet, building on my previous posts, "Does Quantum Mechanics Apply to How People Think" and "Toward a Quantum Model of Humor." It’s a project I’m carrying out with my esteemed colleague Emmanuel Pothos from City University London, and my brilliant Ph.D. student, Iana Bashmakova, who are also co-authors of this particular post.

The Quantum Cognition Approach

The "quantum cognition" approach (as it has come to be called) offers a sophisticated and nuanced approach to the formal description of cognitive processes and phenomena that resist traditional classical models. The approach makes no assumption that phenomena at the quantum level affect the brain but, rather, focuses on the structurally different nature of quantum probability. While in classical probability theory events are drawn from a common sample space, quantum models define states and variables with reference to a context, represented using a basis in a Hilbert space. This results in phenomena such as interference, superposition and entanglement, and ambiguity with respect to the outcome that is resolved with a quantum measurement and the corresponding collapse to a definite state.

The quantum cognition approach has a solid track record for providing solutions to previously unsolved problems in cognitive science (Bruza et al., 2015; White et al., 2015). For example, there is extensive evidence that people use conjunctions and disjunctions of concepts in ways that violate the rules of classical logic (Aerts et al., 2009; Estes & Ward, 2002; Hampton, 1988; Osherson & Smith, 1981). This has made it extremely difficult to provide a mathematical description of how new meanings emerge when people combine concepts and words into larger semantic units such as conjunctions, phrases, or sentences.

However, the quantum cognition approach offers solutions to difficulties arising from classical models of concept combination by using mathematics originally developed for quantum mechanics (Gabora & Aerts, 2002; Aerts & Gabora, 2005a,b; Aerts et al., 2012, 2013, 2016). In addition to modeling concept combination, quantum models can account for phenomena in probability updating (Basieva et al., 2017), individual differences in causal judgments (Mistry et al., 2018), establishing similarity of mental representations (Pothos & Trueblood, 2015), and decision-making (Pothos et al., 2014), which are problematic for models based on classical probability theory. This study is part of a larger effort to (a) understand what cognitive processes or kinds of thinking are particularly in need of a quantum-type framework, and (b) empirically determine whether particular thoughts, mental representations, or ideas are more vs. less quantum-like.

The Question Order Effect

The study we will carry out makes use of a strange yet well-established and replicated effect: the Question Order Effect (QOE). QOE refers to the dependence of question endorsement probabilities on the order of questions (Moore, 2002). For example, if people are first asked how satisfied they are with life in general, and then asked how satisfied they are with specific aspects of their life such as their job and relationships, their answer to the first question reflects their off-the-cuff assessment of life satisfaction in general. However, if the order is reversed, such that they are forced to answer the more specific questions before the more general question, they often answer the general question differently, presumably because they are influenced by having just thought about these specific aspects of their life.

Such order effects are well known in social psychology, but they can be challenging to explain using classical probability theory. By contrast, quantum models can straightforwardly explain QOE. Thus, it is generally accepted that QOE provides a fairly valid and sensitive instrument for diagnosing the degree to which a particular cognitive task requires a quantum-type formalism for its mathematical description (Pothos & Busemeyer, 2022).

QOE and Creativity Tasks

Our study will examine whether the QOE effect is observed not just with simple questions like "How satisfied are you with your life" but also for creativity tasks. The rationale is that creative cognition relies heavily on uncertainty, ambiguity, contextuality, and concept combination (Estes & Ward 2002). In creative tasks, one is confronted with psychological states of ambiguity (i.e., when one has a half-baked idea, one might feel confident one is on the right track by combining A and B, yet be uncertain as to how A and B really go together). Creative tasks also often involve a great deal of contextuality (i.e., it is by reframing a problem, looking at it from a new perspective, that one often comes up with the solution).

Furthermore, in creative tasks, we often have to make inference leaps from some prior to posterior beliefs. In classical probability theory, the jump from priors to posteriors is restricted via Bayes Law, but quantum theory allows stronger leaps (via the quantum equivalent of Bayes Law, Luder’s law; Basieva et al., 2017). Finally, there is theoretical support for quantum creativity models (Gabora, 2017, 2023; Gabora & Carbert, 2015), as well as empirical support for a quantum theory of humor (Gabora & Kitto, 2017), which is a type of creativity that often involves multiple possible interpretations of words or concepts in different contexts.

We hypothesize that there will be a positive relationship between strength of QOE and creativity scores. The coincidence of quantum effects with high creativity would further strengthen the argument that we need that type of mathematics that was originally developed for use in quantum mechanics to describe creative process; in other words, it strengthens the argument for a quantum approach to creativity.

Creativity Essential Reads

Dreaming Up Creativity: The Hypnagogic State and AI

How to Train for the Daily Quest of the Hero Creative

We also have reason to believe that the stimulus novelty of information may promote quantum-like effects. The rationale here is that if a stimulus is novel—i.e., unfamiliar—there may be uncertainty about how to process it. Thus, like creativity, novelty is associated with ambiguity. Also, one has fewer established ways of thinking and reasoning about novel stimuli, so thoughts concerning novel stimuli may be less rigid, and therefore more contextual.

In short, novelty, like creativity, is associated with ambiguity and contextuality, the signature characteristics that make a quantum model potentially applicable. Indeed, novelty, creativity, and quantum effects may tend to go together. Therefore, we have reason to believe that while interaction with familiar concepts may be better approximated with a classic probability model, novel information may require a quantum cognition-type model (similar to Basieva et al., 2017). Thus, there is a second hypothesis tested in this study: not only do we expect there to be a positive relationship between strength of QOE and creativity scores, but we also expect there to be a stronger QOE for unfamiliar stimuli than for familiar ones.