A kind of impalpability to this whole business

What is computational neuroscience?

Computational neuroscience has two faces. On the one hand, it builds computational models of neural phenomenon, analogous to the way computational chemistry, climate science, and computational economics, among others, build computational models of their respective phenomena.  On the other hand, computational neuroscience studies the way the nervous systems compute and process information.  (Piccinini and Shagrir 2014, p. 25)

Our focus is this second facet: Computational neuroscience as the study of how nervous systems perform computations and process information.

If the discipline of cognitive neuroscience is the study of how mental processes are subserved by neural circuits of the brain, then computational neuroscience (as we will use the term) is a theoretical approach to cognitive neuroscience.

Consider the following quotes from the primary literature written several decades ago.

The ultimate aim of computational neuroscience is to explain how electrical and chemical signals are used in the brain to represent and process information. This goal is not new, but much has changed in the last decade. More is known now about the brain because of advances in neuroscience, more computing power is available for performing realistic simulations of neural systems, and new insights are available from the study of simplifying models of large networks of neurons. Brain models are being used to connect the microscopic level accessible by molecular and cellular techniques with the systems level accessible by the study of behavior.  [Sejnowski et al. 1988]

The expression “Computational Neuroscience” reflects the possibility of generating theories of brain function in terms of the information-processing properties of structures that make up nervous systems. It implies we ought to be able to exploit the conceptual and technical resources of computational research to help find explanations of how neural structures achieve their effects, what functions are executed by neural structures, and the nature of representation by states of the nervous system.  [Churchland et al. 1993]

It sounds impressive!  But what exactly do we mean by computation and information in the context of neuroscience?  Is the brain (i.e., the human central nervous system) a computer? Or is this conception an analogy drawn from modern technology that we (perhaps inappropriately) apply to the functional organization of nervous systems?

Jerome Lettvin and other historical characters

Jerome Y. Lettvin, co-author of the classic paper “What the frog’s eye tells the frog’s brain,” was a professor at M.I.T. when I was an undergraduate. Although I only enrolled in one of his classes, General Physiology, I audited his seminar course, Advanced Neurophysiology. Lettvin made an impression on me. I appreciated Lettvin’s hubris, candor, and iconoclastic approach to research and education.

We will visit Jerome Lettvin and his contemporaries — McCulloch, Pitts, and so on — at several points in this course, as we make an effort to understand the history of computational and cognitive neuroscience, as documented in Mind as Machine by Margaret Boden and other sources.  What benefits might there be to emphasizing this historical dimension of computational neuroscience?

The following link will take you to a transcript of Lettvin lecturing on “A view of the function of the neuron,” in which he describes “a kind of impalpability of this whole business” of trying to move from an understanding of neurophysiology to nervous system information processing.   Lettvin insists that the primary difficulties of doing nervous systems research are conceptual rather than technical.  He also insists that the most interesting and vital future scientific work will involve informational as well as material aspects of the world.  David Marr, a well-known computational neuroscientist, writing several decades later, agrees:

Most of the phenomena that are central to us as human beings – the mysteries of life and evolution, of perception and feeling and thought – are primarily phenomenon of information processing, and if we are ever to understand them fully, our thinking must include this perspective.  (David Marr, Vision: A Computational Investigation into the Human Representation and Processing of Visual Information.)

In an effort to reinforce this point, our first videos and readings will be somewhat philosophical in nature.  

Further reading

The Further Reading sections of the course blog will list supplementary papers, videos, books and other resources related to the topic of the day.  These might prove valuable as you think about a term project, or perhaps after graduation.  There are certainly not required reading.

Do not let the Further Reading section(s), which are optional, distract you from reading and writing assignments and preparation class discussion, which are very much required.

If you are a William & Mary student, you can access an electronic version of most of the required and suggested readings under the Readings tab (password protected).