Relational Density Theory: Nonlinearity of Equivalence Relating Examined through Higher-Order Volumetric-Mass-Density

Abstract

We propose relational density theory, as an integration of stimulus equivalence and behavioral momentum theory, to predict the nonlinearity of equivalence responding of verbal humans. Consistent with Newtonian classical mechanics, the theory posits that equivalence networks will demonstrate the higher order properties of density, volume, and mass. That is, networks containing more relations (volume) that are stronger (density) will be more resistant to change (i.e., contain greater mass; mass = volume * density). Data from several equivalence experiments that are not easily interpreted through existing accounts are described in terms of the theory, generating predictable results in most cases. In addition, we put forward the higher-order properties of relational acceleration and gravity, which follow directly from the theory and may inspire future researchers to evaluate the seemingly self-organizing nature of human cognition. Finally, we conclude by describing avenues for real-world translation, considering past research interpreted through relational density theory, and call for basic experimental research to validate and extend core theoretical assumptions.

Merger and expansion of equivalence classes via meaningful stimuli

“…When included in a class of meaningful stimuli, abstract stimuli would become part of a greater network of relations than those generally accessed in experimental preparations. Studies have shown that larger equivalence classes are more stable and more likely to be maintained across time because the network of relations that make up the class become interdependent (e.g., Belisle & Dixon, 2020;Haydu et al, 2009;Saunders, Wachter, & Spradlin, 1988;Spradlin et al, 1992). In other words, the reinforcement of the relation between two class members might strengthen all relations within this class.…”

Improving Fraud Detection via Hierarchical Attention-based Graph Neural Network

“…For all datasets, we use Adam [14] with a learning rate of 5 × 10 −3 for training. For the baselines, we use the open-source implementation 1 . The data format is transformed appropriately to fit their settings.…”

“…Relational Density Theory (RDT) [1] shows that relational mass is associated with relational density and relational volume. Inspired by the theory, we incorporate weighted adjacency matrices across multiple relations to measure tie strength between two nodes.…”

An examination of transformation of evaluative and consequential functions through derived relations with participant-generated values-relevant stimuli

“…One next step in investigating DRR involved in values could involve demonstrating transformation of values functions down a network from a superordinate value to a functional class of goals hierarchically related to that value, to several classes of specific behavioral steps hierarchically related to each valued goal. As another example, an experimental paradigm informed by relational density theory (Belisle & Dixon, 2020) could manipulate the density of values classes and explore the impact on approach and escape responding. Such an approach might allow for a direct assessment of broader patterns of approach behaviors and strengthen the link between basic RFT accounts and mid-level conceptualizations of values in ACT.…”