The most visible systems lesson from virus spreads is related to networks if we know how to see them.
In the early 1980s Faberge Organics Shampoo (the stuff Steve, the character in the Netflix show Stranger Things uses to get his hair so ‘perfect’) ran advertisements that used network theory to serve as a prop for how things — like recommendations for hair care products — spread between people. The movie Wayne’s World even featured a spoof of it. (see below)
This third is a series of lessons on systems thinking and our present-day is not a lesson in public health epidemiology, but on the central feature behind viral spread: networks. More specifically, social networks (and no, it’s not just about Facebook or social media).
Networks: How We Connect
Networks are the central mechanism for understanding human connection at a social scale. Whether it’s small groups, organizations, communities, or societies, interactions in these systems can be partly explained by networks — their structure and function. Understanding network theory can help us explain and affect how, what, where, why and when something spreads — whether it’s a virus, idea, piece of information or some combination.
A social network is loosely defined as more than two things connected for a purpose within a system. These connections are active — meaning there is some exchange — and bounded (we create some limits for where the connection ends and starts based on what the purpose of the connection is).
To illustrate, we might define a social network with boundaries associated with a topic, or function (e.g., information network). These boundaries are ones that can be ‘hard’ (i.e., not adjusted — like all students who are currently enrolled in a specific school on March 20, 2020) or ‘soft’ (i.e., adjusted — such as all members of a neighbourhood Facebook discussion group on a topic where members join, leave, and where neighbourhood isn’t clearly defined to a specific geographic boundary).
Why does this matter?
The boundaries we set (or move) and the purpose we identify determines what we see. What we see determines what can track, manipulate, and design for. COVID-19 is a novel coronavirus. When the outbreak first started we didn’t track it because we didn’t know it existed. Quickly that changed.
People didn’t distance, because they weren’t tracking it. The stock markets didn’t react right away because they didn’t track social fear, investor confidence, and more.
You get the idea.
Networks: Why We Connect
We, humans, are defined much by our social connections. Whether it’s for survival, comfort, or identity — so much of who we are is wrapped up in our connections. Network theory explains some of these connections and helps us to understand how to approach network ‘problems’.
Homophily reflects the aphorism “birds of a feather, flock together“. It refers to those with similar attributes or qualities. We like people who are like us. Groups that are similar are efficient, can more easily rely on shared heuristics for decision-making, and tend to aggregate easier because the ‘social distance’ is low. The downside with such network structures is that the similarity is also what makes them more susceptible to shocks and Groupthink (the blind spots created through an entrained perspective or culture).
Balance is a theoretical perspective that seeks to build on our similarities and extend our advantages. It’s about ‘friends of friends‘ with an assumption that a friend of yours is likely to have qualities similar to you and therefore is likely to be similar to me. This is how we extend our networks. Our friends are likely to be similar to us and friends of friends a little less so.
Structural Holes are theories about how we bridge gaps. The line “you complete me” from the film Jerry MacGuire reflects this theory. Or, for those of a certain age and culture: your peanut butter needs my chocolate. Structural holes are about creating connections based on the need to fill a hole by connecting two things that create a whole.
When We Connect
Network theories also help us to recognize the advantages of certain structures and patterns of transmission across those networks. For example, theories of proximity look at how close — literally or figuratively — to each other. The closer we are, the more likely we are to do things like share attributes, information, experiences and co-learn.
This is being put to the test on a massive scale with social distancing and remote work where the usual physical proximity we’re accustomed to is eliminated from many circumstances. Here is where understanding the benefits and patterns of communication associated with proximity matter. Seth Godin’s Forward Link network has created a virtual water cooler and co-working ‘space’ for people (anyone — it is free) to have the kind of impromptu conversations that they normally might have at their offices or cafes. This recognizes the value of proximity.
Theories of co-evolution look at how we evolve our practices together, particularly in complex systems. Right now many of us are learning how to use tools like Zoom or Skype as teaching platforms. Visual communication ools like Mural, Miro, and Milanote allow for virtual collaboration and visualizing just like in person (but with obvious differences). For example, Milanote provides a visual interface for sharing and collecting artifacts that allow us to transcend the distances created by working remotely to do what many would do in person: spread out, share, discuss, edit, and recombine ideas. In times of uncertainty, complexity, physical distance, and need for innovative thinking and action tools like this are vital.
Many people are learning this together and will evolve their practices to adapt to the situation and the tools. This is how we change the culture.
Theories of collective action are network theories that examine what happens when we coordinate and act on a certain scale. This also introduces the popular (but often misunderstood) concept of critical mass. These theories examine both the effects of coordination, but the requirements necessary to make something happen. It’s not enough to have people work together in some cases, there must be a certain amount of activity at a certain scale to influence change. Critical mass is (partly) an assessment of that scale. The number of people required to act together to effectively contain the spread of something is one such assessment.
What to do?
Networks are all around us. They are being built, falling apart, and evolving all the time and right now it feels like it’s all happening very fast. There is no such thing as a ‘good’ social network model. Each of the perspectives brings advantages and disadvantages to a situation. Networks like the one illustrated below of members of a professional practice network can also have different sub-structures within them.
What you’ll see at the centre of this network map are a few people (represented by boxes) densely connected with many others (indicated by the number of lines to other shapes — leave aside the matter of line thickness for now). What you will see from this is a variety of sub-networks in the group.
For example, look at the person identified as 5208 (pink, near the centre). That person is connected to a number of people who are similar to them in circumstance (shape), but less in role/setting (colour). They have a large number of connections to others and many of those others are connected to a further group of people.
What about 6306 (green triangle, near the bottom). This person is connected to many others, but many of those aren’t connected to anyone else.
Comparing the two people together we see very different roles and structures. For information flow, 6306 is a highly efficient and powerful conduit for distribution and exchange. However, if that person is no longer active or engaged in the network, much gets lost because many connections are lost.
5208 is different. If they leave, most of the connections to the network from others aren’t significantly affected. They are a slower, but more robust connection.
If you need to get a message out or spread an innovation quickly, hub-and-spoke models like those with 6306 are better-suited. For slower, but more robust networks, the kind that 5208 is a part of is a more suitable structure. What’s further is that the role of each player or node might change as the network changes and evolves.
Networks By Design
The network above was partly intentional. We can design our networks to suit our needs. If speed or efficiency is required, nurturing hubs makes sense. For vulnerable networks, we can promote ways to get people connected to others, preferably to those who are different than they are. Collectively, we can create networks that look somewhat like what we see above: include a mix of many different types of networks.
Most social networks combine an element of intentional strategy and happenstance. We can change the mix with design — if we’re intentional about it.
Be safe, connect, and do so wisely.
Networks don’t just ‘happen’; they are often designed. If that’s what you’re looking to learn through networks and create, evaluate, and support their development — contact me through Cense Ltd (that’s what we do).
Note: An earlier version of this was originally shared on Censemaking.com