Real Patterns in Science and Nature: Q&A with Assistant Professor of Practice Tyler Millhouse

What makes a pattern real? More than three decades ago, philosopher and cognitive scientist Daniel C. Dennett posed that deceptively simple question in his landmark 1991 essay “Real Patterns.” Today, as artificial intelligence reshapes how researchers detect, model and interpret information, the question feels newly urgent.

In Dennett’s Real Patterns in Science and Nature (The MIT Press, 2026), College of Information Science Assistant Professor of Practice Tyler Millhouse and co-editors Stephen Peterson and Don Ross revisit Dennett’s influential idea and trace its implications across science, philosophy and AI. The volume brings together leading thinkers to examine how patterns give rise to meaning, explanation and understanding—from the emergence of structure in physics and the nature of biological species to economic welfare, causal reasoning and the possibility of genuine understanding in large neural networks.

The result is both a tribute to Dennett’s enduring influence and a timely exploration of one of the central questions of the information age: How do patterns help us make sense of the world?

In this Q&A, co-editor Tyler Millhouse discusses the origins of Dennett’s Real Patterns in Science and Nature, what readers will take away from the book, how it connects to information science and how his own research shaped the project.

What prompted this project and did you experience any setbacks along the way?

The book project began with a conference I helped to organize (along with my fellow editors) at the Santa Fe Institute in 2022. It took a while to get the volume to publication due to a few major setbacks. The biggest of these was that Dan Dennett (who had originally planned to write a short response to each chapter) passed away unexpectedly. A number of the editors and authors (myself included) knew him personally, so this was a significant blow. Fortunately, we were able to pivot and frame the volume as a well-deserved tribute to his work.

Who is the target audience and what do you hope readers take away from the book?

The target audience of the book is anyone with an academic interest in what science tells us about reality. The one big idea of the book is real patterns—an idea first proposed by Dan Dennett in his 1991 paper "Real Patterns." In this paper, Dennett suggests that much of what the various sciences talk about (from species and star clusters to markets and mental states) can be understood as patterns in the information theoretic sense. Crucially, information theory proposes criteria for assessing whether patterns are present in data, and Dennett adapts these criteria to assess whether patterns are present in the world.

Dennett’s Real Patterns in Science and Nature is a survey of different variations on and applications of this basic idea, including applications in physics, biology, economics, linguistics and cognitive science. If anyone is interested in real patterns but not sure where to begin, the co-editors and I have written a detailed and accessible introduction as the first chapter and reprinted the original paper as the second chapter.
  
  

Many conclude that all scientific models are, to some extent, useful fictions. On the contrary, real patterns proposes that information theory can help us to understand in what sense imperfect, limited or idealized models really do tell us about reality.

  
Can you tell us a bit more about how the book fits into the field of information science?

The entire book concerns how information theoretic accounts of patterns can help us to better understand what science is telling us about the world. There are a number of concerns, for example, about what if anything scientific models tell us about the world. Scientific models exhibit varying degrees of simplicity, predictive accuracy, explanatory scope and idealization. This has led many to conclude that all scientific models are, to some extent, useful fictions. On the contrary, real patterns proposes that information theory can help us to understand in what sense imperfect, limited or idealized models really do tell us about reality. 

How does the book fit into your research and, more broadly, reflect upon the College of Information Science?

While I admit to being a little biased, I think the book reflects very well on the college. The MIT Press is one of the top publishers for academic work in philosophy (especially in this area). We've also gotten many of the top scholars on the subject to contribute. For example, David Wallace and Sean Carroll (also a famous science communicator) are two of the leading figures in the philosophy of physics.

More personally, the book is both about my research and contains my research (I contributed a chapter and am the first author of the introduction). There has been rapid growth in publications on real patterns in the last decade, and I hope to further promote this area of research, which will in turn create more opportunities to publish in this area. With respect to teaching, I have found that I continually return to the core ideas of the book to explain concepts in my artificial intelligence course. One of the main topics in the course is how to select the right AI tool for the job, and whether a tool "fits" a task depends a lot on the kinds of patterns we want to capture or utilize and whether those patterns are present in our target domain.
  

If you like the book cover, Tyler has built this interactive web tool so you can make your own images in the same style! 
  

Assistant Professor of Practice Tyler Millhouse specializes in the philosophy of science, with an emphasis on the philosophy of data science, machine learning and artificial intelligence. He recently completed a postdoctoral fellowship at the Santa Fe Institute under the direction of Melanie Mitchell. His research has appeared in leading journals, including the Australasian Journal of Philosophy, The British Journal for the Philosophy of Science and Philosophy of Science. He holds a BA in Philosophy from Ashland University, MA in Philosophy from Tufts University, and PhD in Philosophy from the University of Arizona.