Skip to main content

Diffusion and Movie Choices

Here at Cornell I am a part of several different friend groups that are relatively unconnected (I don’t think I am alone in this). I have my friends from freshman year, my friends from my project team, and my friends from my major/classes. These groups all overlap by one or two people, but by and large they do not overlap. Similarly, I know that many of the friends I have in each of these groups also have their own separate groups that I am not a part of. This is analogous to what we have been talking about in class lately about clusters. Each of my friend groups is a very dense cluster. We are all friends with each other within each group, but there are few edges connecting the groups besides my connection to each of them.

Lately I observed a phenomenon that reminded me very closely of the model we have been discussing about diffusion and cascades in networks. As discussed in this article ( from Psychology Today, picking a movie can be challenging. There is so much conflicting information out there about movies. Two of my friend groups were both talking about going to see a movie on Friday night, but they each wanted to see different movies. Let’s call one of these movies A and the other B. The movies were both reputable movies with good reviews (one of the main sources of information that the article lists for people when choosing their movie). However, this article largely ignores what I believe to be one of the main factors that influence movie choices: the preferences of friends. I wanted to see movie A. I had the option of just going with the friend group that wanted to see this movie, but I wanted to be able to spend time with all of my friends. But, being the one edge that connected the two groups, I was the only one trying to convince everyone else in the other group that we should see movie B, so I was unable to convince anyone. Now that I have the insight that I have about networks, I can see why. The densities of the groups were so high that the threshold to switch was simply also too high for anyone in the second group to be convinced by only me, so I was unable to spread the cascade all the way through the network. The quality of the reviews could be related to the threshold for switching. If one movie has excellent reviews and the other doesn’t, maybe people would be more likely to switch to the one with good reviews. However, this process is definitely driven largely by diffusion of preferences in networks. They went to see movie B and we went to see movie A, and everyone had a good time.


Leave a Reply

Blogging Calendar

November 2018
« Oct   Dec »