The Relative Contributions of Traits and Contexts On Social Network Learning

Understanding how people are embedded in social networks helps us navigate the social world. Yet, we learn and remember people's social attributes differently. Furthermore, it's unclear if distinct memory mechanisms for different social attributes might distort our representations of these social networks. Here, we study the relative contributions of shared contexts and personality traits when learning friendships in a social network. Specifically, we were interested in whether trait valence or specific contexts shape how humans remember friendships in social networks. Through varying levels of network topological complexity, we find the contexts where people know each other are most memorable and that better contextual retrieval uniquely relates to better recall of friendships. Moreover, subsequent computational modeling revealed that participants' biases for encoding a network member's personality traits and network position has a downstream impact on friendship recall. These results suggest contextual memory can serve as a scaffold for mapping out dyadic friendship relationships in social networks, and illustrates the acute nature by which a person's personality traits impact memory of their respective friendships. More generally, our findings give insights into how the same social network can be represented differently depending on individual learning biases and experience.