Modeling Memory Imprints Induced by Interactions in Social Networks

James Flamino, Ross DeVito, Omar Lizardo, Boleslaw K. Szymanski

Abstract: Memory imprints of the significance of relationships evolve over time. This evolution is driven by these imprints, gaining strength during interactions between the people involved, and weakening between such events. Despite the importance of understanding this evolution, few research papers explore how long-term interactions in social networks correlate with the memory imprints of relationship importance. In this paper, we represent memory dynamics by adapting a well-known cognitive science model. Using two unique longitudinal datasets, we fit the model’s parameters to maximize agreement of the memory imprints of relationship strengths of a node predicted from call detail records with the ground-truth list of relationships of this node ordered by their strength. We find that this model, trained on one population, predicts not only on this population but also on a different one, suggesting the universality of memory imprints of social interactions among unrelated individuals. This paper lays the foundation for studying the modeling of social interactions as memory imprints, and its potential use as an unobtrusive tool to early detection of individuals with memory malfunctions.