South Australian Tall Poppy Award

Social network theory

When he was growing up, Lewis Mitchell wanted to become a scientist. In fact, he quite liked the idea of a career in theoretical physics. He didn't know what it entailed, but it sounded cool.

"I wanted to create my own universes and play in those," he says. "But then I hit university and discovered it wasn't all science fiction – you had to do lab work as well."

Laboratory life wasn't his cup of tea, but the mathematics coursework intrigued him.

"In applied mathematics, you do a lot of simulations on computers," says Lewis. "I realised I could do all the fun stuff I wanted to do from the comfort of my own computer!'"

Soon, he was performing complex statistics related to numerical weather prediction, including ocean wave simulations. This work took him overseas, where his career took another interesting turn.  

"During my post-doc in the US, I was surrounded by people who were studying Twitter," he says. "I ended up talking to those people a lot."

These social interactions in real life led to a profound interest in studying social interactions online.

Today, Lewis is a Senior Lecturer in Applied Mathematics at The University of Adelaide, and an Associate Investigator for ACEMS, where he develops statistical tools to monitor population-level trends in online social media in order to understand how ideas spread on the internet.

"You hear a lot about echo-chambers and filter bubbles. We're trying to find out how those things actually form," says Lewis.

"In particular, I'd like to understand how algorithms on different social media platforms promote different types of discussion."

By generating mathematical models of online information flow and then comparing these with 'real world' observations, he and his colleagues are gaining a better understanding of how both information and misinformation spread, and the important role local networks play in this.

His findings show that the people you're in contact most often are the ones you have the most influence over. He believes this has important implications for science communication.

"Scientists have an important role to play as 'trusted nodes' in their local communities and local parts of the network," he explains.

This helps spread correct scientific information as well as trust in science.

"This is something ACEMS does really well," he adds. "It provides opportunities for us as scientists to get into our local communities and propagate good science there." 

Lewis’ childhood love of science continues to grow. Ultimately, he wants to create algorithms that promote healthy discussion and propagate constructive information.

That would certainly make a big difference in our local universe.

In particular, I'd like to understand how algorithms on different social media platforms promote different types of discussion.