In the past, a person might be diagnosed with hysteria — a mental condition for which no useful treatment could be possible, because the diagnostic category was too broad and unfounded. The idea of “hysteria” limited our understanding of the human mind. SFI Complexity Postdoctoral Fellow Marina Dubova wants to challenge contemporary categories and methods in science that may limit rather than enhance our understanding of the world.
“Many principles underlying the scientific method, such as the preference for simpler explanations (Occam’s razor) or the focus on rigorous testing of existing theories over exploratory experimentation, are primarily based on intuition,” she says. “Today, we have empirical methods and results that can provide evidence-based insights into these questions, exploring which approaches best facilitate scientists’ ability to learn about the world.”
Dubova holds a Ph.D. in cognitive science from Indiana University, and was recently a graduate visitor at Carnegie Mellon University. Dubova’s research findings contend that experimenting more openly, even randomly, can help scientists develop more accurate theories than those based on previous assumptions. She has established an evidentiary basis for complexity preference in theory-building — as opposed to simplification and parsimony — leading to more efficient learning and more accurate results. “Rather than compressing rich phenomena into discrete dimensions, maybe there are ways of theorizing that are expansive, instead of reductive,” says Dubova.
An alienist from the 1800s might have found the concept of hysteria easy to conceive, but the consensus now is that such a category did not represent truth in useful or accurate ways. At SFI, Dubova plans to use cognitive methods and insights to rethink the key methods that can help scientists learn about the world — including experimentation and theory-building methods and ways to refine scientific conceptual systems — that leave oversimplified explanations behind. She begins her fellowship in August.