Demystifying the Irrational: A Neural Network Approach to Understanding Superstition Through Subjective Norms and Anxiety

Abstract

This study aims to address the gap in the literature on superstition by testing the interplay between subjective norms and anxiety as predictors of superstitious beliefs and behaviors. We tested the interaction effect between anxiety and subjective norms in a sample of 110 participants. The data indicates that trait anxiety accentuates the positive association between subjective norms and superstition. Moreover, in order to broaden the understanding of this interaction, a neural network was developed and trained using the real survey data. After training the neural network with the real data, we simulated extreme levels of anxiety in order to predict superstition. Results reveal that subjective norms significantly predict superstitious beliefs, while anxiety has a lesser main effect. Additionally, subjective norms have a stronger association with superstition when anxiety levels are high rather than low. These simulation results allow us to make causal claims about the effect of subjective norms and anxiety on emergent superstitious beliefs. They also contribute to the understanding of the cognitive mechanisms that explain the emergence of superstitious behaviors and beliefs.