In designing our social network, we will consider how we intend to facilitate communication between individuals. One important consideration is: who will be communicating to who? This post will identify who people will interact with and why.
One way we can study this is by considering the models of homophily and heterophily. Rogers and Bhowmik (1958) discuss the concepts of homophily and heterophily with regard to human communication. They define the concepts as follows:
“Homophily refers to the degree to which pairs of individuals who interact are similar with respect to certain attributes, such as beliefs, values, education, social status etc.”
“Heterophily is the degree to which pairs of individuals who interact are different with respect to certain attributes.”
We’ll discuss these two models below.
Rogers and Bhowmik explain that, when a person is presented with a number of choices of individuals to interact with, there will be a tendency to choose individuals who are like themselves. They further suggest that this is because “more effective communication occurs when source and receiver are homophilous”.
Rogers and Bhowmik assert that heterophilous communication is typically ineffective, driven by empathy of the individuals involved. The empathy is needed to project themselves into the role of another person. Rogers and Bhowmik suggests that heterophilous communication is far more effective when the source has greater empathy than the receiver.
Homophily in our social network
In our social network, we wish to encourage the homophily relationship, therefore encouraging groups of individuals with similar interests to communicate and share experiences. This can be seen in the concept of “rooms”, which encourage the discussion of particular topics by scanning that subject in real life. Individuals within the room will have also scanned this item, so will presumably have similar interests.
Rogers, E. M., & Bhowmik, D. K. (1970). Homophily-heterophily: Relational concepts for communication research. Public Opinion Quarterly, 34(4), 523-538.