Death can, for obvious reasons, have a profound impact on social networks. When someone dies, it can cut off ties between people who only knew each other through the decedent, for one thing, and it can cause other relationships to grow closer as members of the network cope with the loss. What’s interesting, and a little bit weird, about life in 2017 is that we can watch some of these dynamics unfold online. Online social networks aren’t identical to offline ones, of course, but researchers do think that in certain important ways they act similarly.
In a new research letter published in Nature, William R. Hobbs of UC San Diego and Moira K. Burke, a Facebook researcher, explain the results of a study they conducted geared at better understanding what happens to an online social network after someone dies. To oversimplify a bit, they basically analyzed a very large number of Facebook interactions in different social networks (from anonymized data) before and after the deaths of members. That allowed them to compare the interactions they saw after the death with the interactions they could have expected to have occurred had no death occurred, and it allowed them to break down differences in effects between close friends, acquaintances, and strangers in the same networks.
Summing up their findings, Hobbs and Burke write that, “As expected, a substantial amount of social interaction was lost with the death of a friend. However, friends of the decedent immediately increased interactions with each other and maintained these added interactions for years after the loss.”
It’s an effect nicely summed up by this before-and-after graph — you can ignore the technical language in the caption — in which the vertical line in the middle representing the point in time at which the friend dies.
As you can see, the death really does seem to spur an increase in online social interactions in all the groups other than those who were already close friends at the time of the loss — one which lasts for a long time.
Now, there are always going to be limitations to the study of online versus offline networks, and the authors do note that “because we do not have data on offline interactions, we cannot say for certain that social support online reflected increased offline interactions.” But these are still interesting findings, and there is something at least vaguely comforting about the idea of someone’s death bringing people in his or her orbit a bit closer for years to come.