A map created using Quid showing news articles about neuroscience discoveries made possible with DTI (Diffusion Tensor Imaging). DTI reveals tracts of white matter connectivity in the brain, allowing us to see which regions of the brain are talking to one another.
An Ecosystem of DTI Discoveries
About the map: Each of those colored dots is one news story. Dots are called nodes. They are connected by lines called edges. Using natural language processing, algorithms can read the text content of articles and assess content. What is the article about? What are the key concepts? This information creates a similarity matrix of sorts, describing nodes by attributes. These attributes define the physical layout of the network. Articles most closely linked are grouped in clusters called communities.
This map shows an initial set of 550 nodes arranged into network view. Each node repels other nodes, making the communities with fewer connections move farther apart and those with more connections cluster together. It’s all generated in Quid.
A network layout is just the first step. For it to be useful, I need to understand it.
Next comes my favorite: human exploration.
It takes work to make sense of a complex ecosystem of information. Here’s my technique to tackle a network: Make a first pass, exploring each cluster. As I slowly begin to familiarize myself with the graph, I give each community a name.The cluster names are added manually, so it’s helpful to explore the largest nodes and nodes that stand out. Along the way, I save articles of interest.
Once each community is named, it’s easier to dive into details that make a big picture – the individual stories of scientific discoveries. I particularly love looking at stories that bridge two or more distinct communities.
Now that I’ve collected links, it’s time to embark upon digesting 20+ articles and abstracts spanning things like white matter connectivity associated with self esteem, the neuroscience of risk-taking, and even the links between physical fitness and brain health. Stay tuned for my completed thoughts next month in Scientific American and hit me up on Twitter if this post gave you any ideas!