Infection and Information Diffuse Differently
Although many phenomena exhibit diffusion across networks, this diffusion may happen very differently depending on what is diffusing. Even within the study of the spread of disease across networks, it is important not only to look at how infection spreads, but also how beliefs and information spread across these same networks. Beliefs about the effectiveness of immunization and healthcare and information about the severity of a specific disease, its level of contagiousness, and how it spreads can all influence behaviors of people in a network. These behaviors can, in turn, have an impact on a person’s risk for contracting a disease. Thus, the spread of disease is influenced by the network structure, but also by the spread of beliefs and information through that same network structure. Understanding the network structure is important for understanding both the spread of infection and the spread of beliefs, which may itself influence the spread of infection.
Following this logic, Grim, Reade, Singer, Fisher, and Majewicz studied both the spread of information and the spread of disease through networks with different structures. They focused specifically on the dynamics within sub-networks, acknowledging that social networks have many clusters with different structures and analyzing the importance of bridges, or links between sub-networks. The study was published in the December 2010 issue of the journal Connections, which is available here. (Their article, “What You Believe Travels Differently: Information and Infection Dynamics Across Sub-Networks,” begins on page 50.) In their study, they found that network structure has a strong influence on the spread of infection. In fact, they claim that the structure is of primary importance in determining how and how quickly disease will spread, with the degree of linkage between sub-networks having only a minor importance. On the other hand, the degree of linkage between sub-networks is of primary importance for the spread of beliefs, regardless of network structure.
Specifically, Grim, et al. found that the spread of information and beliefs follows the pattern of a power law. This is the same general pattern exhibited by the distribution of popularity for things such as books, songs, and clothing, as explained in class. This suggests that reaching a consensus of beliefs and information can be thought of as such beliefs becoming “popular.” As noted by Grim, et al., understanding the spread of beliefs and information about diseases is important to design effective strategies to prevent and combat the spread of the disease itself. Based on the results of this study, public health organizations and officials should target especially isolated groups to encourage vaccination and spread accurate information. They should also target bridges between sub-networks, since the existence and quanitity of these nodes have a great impact on the speed with which information spreads. They should worry less about highly interconnected sub-networks because information will spread quickly to and through these groups, especially once it has spread to the bridges.
This study is interesting because it focuses on multiple levels of network dynamics. It analyzes the spread of both information and disease, and distinguishes the dynamics by which each is characterized. This is especially interesting because the spread of information about a disease can impact the spread of the disease itself, and thus the two separate patterns may have overlapping influences. Furthermore, the study analyzed not only the type of structure of networks but the role of bridges and of the degree of connectedness within and among sub-networks. This study builds on what we’ve done in class by comparing the patterns of different diffusions through networks, determining which structural factors influence each pattern of diffusion. Furthermore, the authors note that multiple things can diffuse through networks simultaneously, with each pattern of diffusion having possible consequences for another. Investigating the simultaneous spread of disease and information, and the overlapping dynamics between the two, could provide very useful data about public health crises and interventions and could be very helpful for designing strategies to combat the spread of diseases.