Applied and Conceptual Implications of Relational Density Theory in Machine Learning and Behavioral Economics

Relational Density Theory (Belisle & Dixon, 2020a) is an advanced conceptualization of Stimulus Equivalence (Sidman & Tailby, 1982) and Relational Frame Theory (Hayes et al., 2001) that allows behavior scientists to measure the strength, formation, and malleability of covert behavioral experiences such as attitudes, biases, and probabilities. Relational density combines principles of behavior with the fundamentals of Newtonian classical mechanics to quantify the dimensions of relational framing as a tool for examining behavior. This thesis contains two studies on relational density in context. The first study utilized relational density theory and principles of behavioral economics to measure attitudes and charitable response allocation towards members of the unhoused community. Results indicated emotional relatedness and charitable probability were altered as a function of relational frames and degree of social relatedness. The second study tested for relational density consistent effects in artificial neural networks, a division of machine learning. Results demonstrated the potential for further research into the behavior analysis of machine learning using relational density paradigms.