Models of the Mind

Models of the Mind: How Physics, Engineering and Mathematics Have Shaped Our Understanding of the Brain (Bloomsbury, 2021) provides a multifaceted and approachable introduction to theoretical neuroscience. It discusses some major topics of the field, including both the milestones from their history and the currently open questions.

It's accessible for a general audience, not expecting any previous knowledge of neuroscience or maths. At the same time, neuroscientists have described it as impressive. According to Gaute Einevoll, professor of brain physics, "this is a book that belongs on the bookshelf of any computational neuroscientist and lots of other people".

The first chapter explores the role of mathematical models in biology. It emphasizes that a model can't capture all important aspects of a complex biological system (like the brain). At the same time, if we want to understand such a complex system, we need models that can describe and perhaps even quantify a particular aspect of it.

"Mathematical models of the mind do not make for perfect replicas of the brain, nor should we strive for them to be. Yet in the study of the most complex object in the known universe, mathematical models are not just useful but absolutely essential. The brain will not be understood through words alone."

The following chapters present several examples where a model borrowed from physics, engineering, or mathematics has helped in understanding a particular facet of the brain. Chapter 2 focuses on electricity, which is a major factor in the communication between neurons. At the beginning of the 20th century, some scientists started to describe neurons in terms of an electrical circuit. This analogy has turned out to be hugely successful. Pondering what might play the role of a resistor or a capacitor has led to various insights about neurons' cellular components.

Since the middle of the 20th century, neuroscience and the even newer discipline of computer science have constantly influenced each other. Models of computation demonstrated what networks of connected neurons firing a signal in a yes/no fashion are capable of. They can do complex computations, store memories, combine the inputs of several excitatory and inhibitory neurons. Chapters 3-5 describe the interplay between computer science and neuroscience in these areas.

Chapter 6 is dedicated to vision, where the development of neurobiology and AI has been "uniquely interwoven". Engineers building image recognition systems have used biological vision as inspiration. And AI systems have given important clues for neurobiologists studying vision.

Graph theory is a subfield of mathematics that describes the relationships between different components. Chapter 9 provides many examples of how these kinds of models have been helpful in neuroscience, analyzing which neurons are connected to each other within a smaller or bigger network. It also points out which questions these connectome models can't answer.

A famous "cautionary tale" is about a group of 30 neurons that guide the digestion in a lobster. The 195 connections in this network were mapped out in the 1980s. In the 1990s, Eve Marder and her group showed that the same network can produce very different rhythms depending on which neuromodulator chemicals are present. At the same time, they also showed that different structures can produce the same behavior. Both insights have had implications far beyond the lobster's digestion. Marder's work demonstrated that detailed structural information is necessary but not sufficient to understand how a neural circuit (even such a small one) works.

Models of the Mind does a great job not just in presenting the most important discoveries of the field, but also in discussing the unanswered questions. Chapter 5 explains how oscillations can be observed and measured (!) in many parts of the brain. It also gives an overview of the debate among neuroscientists about the importance of oscillations.

Chapter 7 tells how information theory has inspired the search for a "neural code". Various coding schemes have managed to explain some specific circuits, but definitely not "the brain" as a whole.

"For the question of why the neural code is such an enigma, the most likely answer - as with so many questions of the brain - is because it's complicated. Some neurons, in some areas of the brain, under some circumstances, may be using a rate-based code. Other neurons, in other times and places, may be using a code based on the timing of spikes, or the time in between spikes, or some other code altogether."

Chapter 8 explores the motor cortex, one of the most mysterious parts of the brain. Again, we can see several examples where science hasn't reached a clear answer, but has gained a lot of insight. "Just because 'what does the motor cortex encode?' was the wrong question to ask to understand the motor cortex doesn't mean the answer has no value."

In the pursuit of explaining neuroscience, Models of the Mind also provides excellent expositions of concepts that are widely used in multiple disciplines. Think about terms like "chaos", "probability", or "p value".

In our conversation, we covered some of the overarching themes of the book. The constant push and pull between mathematics and biology: mathematical models simplifying complex phenomena and biology pointing out the importance of a specific detail. What efficiency means for a biological system, like the brain. Whether and how much we can assume that an evolved system is efficient.

Dr. Grace Lindsay also talked about how science communication has helped her explore and discuss topics not directly related to her research. She started blogging and podcasting during her PhD, which has led to further writing opportunities, including this popular science book.

Similar to Models of the Mind, the Lindsay Lab is multidisciplinary: It uses artificial neural networks for psychology, neuroscience, and climate change. In the interview, Dr. Grace Lindsay talked about her decision about the lab's profile She explains the overlap in technologies used for studying visual systems and satellite images. We also hear about examples of how scientists in various fields have taken on research topics related to climate change.

Links: 

• Dr. Grace Lindsay's homepage
• Lindsay Lab
• Dr. Grace Lindsay's blog post about Models of the Mind