From Uncertainty to Insight: An Autocatalytic Framework

Abstract

We show how uncertainty and insight can be modeled using Reflexively Autocatalytic Foodset-generated (RAF) networks. RAF networks have been used to model the self-organization of adaptive networks associated with the origin and early evolution both biological life, and the kind of cognitive structure necessary for cultural evolution. The RAF approach is applicable in these seemingly disparate cases because it provides a theoretical framework for formally describing systems composed of elements that interact to form new elements, and for studying under what conditions these (initial + new) elements collectively become integrated wholes of various types. Here, the elements are mental representations, and the whole is a conceptual network. The initial components-referred to as foodset items-are mental representations that are innate, or were acquired through social learning or individual learning (of pre-existing information). The new elements-referred to as foodset-derived items-are mental representations that result from creative thought (resulting in new information). The demarcation into foodset versus foodset-derived elements provides a natural means of (i) grounding abstract concepts in direct experiences (foodset-derived elements emerge through 'reactions' that can be traced back to foodset items), and (ii) precisely describing and tracking how new ideas emerge from earlier ones. Thus, RAFs can model how endogenous conceptual restructuring results in new conduits by which uncertainties can be resolved and needs can be met. A source of uncertainty is modeled as an element that resists integration into the conceptual network, which is described as a maxRAF containing the bulk of the individual's mental representations. This disconnect produces arousal, which catalyzes one or more interactions amongst mental representations. We illustrate the approach using the historical example of Kekulé's realization that benzene is ring-shaped through a reverie of a snake biting its tail. We show how a single conceptual change can precipitate a cascade of reiterated cognitive 'reactions' (self-organized criticality) that affect the network's global structure, and discuss why this may help explain why cognitive restructuring can be therapeutic. Finally, we discuss educational implications of the RAF approach.

An Autocatalytic Network Model of Conceptual Change

“…Existing conceptual structure increasingly constrains and enables the scaffolding of new conceptual structure. Reactions at one layer (e.g., factual knowledge, or phonological similarities) may spark chain reactions that lead to activation at other layers (e.g., metacognitive knowledge), a phenomenon discussed elsewhere in relation to selforganized criticality and insight (Gabora & Steel, 2021b). Thus, individual RAFs form and merge into an expanding maxRAF, which eventually subsumes most of the developing child's conceptual structure.…”

“…It has been proposed that the origin of a self-modifying autocatalytic cognitive network is the key to cumulative cultural evolution in our species (Gabora & Steel, 2020b), and the source of our generativity (Gabora & Steel, 2021b). Since autocatalytic networks incorporate the "triggering" or "catalytic" effect that an observation or piece of information can have on network structure, they provide a means of incorporating creativity into cognitive networks.…”

“…Such generativity may result in phase transitions to a network that is self‐sustaining and self‐organizing (Hordijk & Steel, 2004, 2016; Mossel & Steel, 2005), as well as potentially able to self‐replicate (in a relatively haphazard manner, without reliance on a self‐assembly code), and evolve, that is, exhibit cumulative, adaptive, open‐ended change (Gabora & Steel, 2021a; Hordijk & Steel, 2015). Therefore, RAFs have been used to model the origins of evolutionary processes, both biological—the origin of life (OOL) (Hordijk, Hein, & Steel, 2010; Xavier et al., 2020)—and cultural—the origin of culture (OOC), or more specifically, the kind of cognitive structure capable of generating cumulative, adaptive, open‐ended innovation (Gabora & Steel, 2017, 2020a, 2020b, 2021b).…”

Exploratory exploitation and exploitative exploration: The phenomenology of play and the computational dynamics of search

“…The separation of generative activation and evaluation in NSM stems from adopting the view that creators generate as many solutions as possible and then choose the best, a view that is inconsistent with the results of studies of analogy making and artmaking (Carbert, Gabora, Schwartz, & Ranjan, 2014;Gabora & Saab, 2011;Scotney, Schwartz, Carbert, Adam Saab, & Gabora, 2020). Even when a creator generates multiple possibilities (e.g., multiple sketches for a painting) and chooses one, these seemingly distinct possibilities may be just different ways of expressing a single underlying idea that the creator is wrestling with (Gabora, 2019;Gabora & Steel, 2022), and in so doing, forging their personal creative style (Gabora, O'Connor, & Ranjan, 2012). These different possible expressions of an ill-defined idea have been modeled as different projections of a superposition state (Gabora, 2017;Gabora & Carbert, 2015).…”

“…In this view, the underlying cognitive process is the same in both; that is, in both one is looking at something in a different context. However, the unconventional contexts considered in DT result in widely different conceptions (which get counted as different ideas), while the similar contexts considered in CT result in similar conceptions (which get counted as refinements of the same idea) (Gabora, 2019;Gabora & Steel, 2022;Scotney et al, 2020). (A useful analogy is: DT is like shining light on an object from very different angles, producing differently shaped shadows, while CT is like shining the light from similar angles, producing similar-shaped shadows.)…”