The Financial Times and its journalism are subject to a self-regulation regime under the FT Editorial Code of Practice.
Last updated: June 2, 2012 12:04 am
Connectome: How the Brain’s Wiring Makes Us Who We Are, by Sebastian Seung, Allen Lane, RRP£20/Houghton Mifflin Harcourt, RRP$27, 384 pages
Be prepared for the information about to enter your brain via your retina to cause neuronal firings that will result in the raising of your eyebrows: a leading neuroscientist has declared that that we are not mere collections of atoms, nor just biological machines. According to Sebastian Seung, the self is a non-material entity. The 21st-century soul, however, is not some strange ghost in the machine. It’s information. You are, in effect, a piece of software running on the wetware in your skull.
It’s been clear for some time that our existence as conscious individuals depends on the functioning of our brains. But our individuality is not simply a matter of biological identity. You are born with pretty much all the neurons you will ever have, yet a human with the same brain, built by the same DNA, would become a very different person if raised in the Amazon rainforest or in south London.
What changes over time, and what forms you into the person you become, are the connections between these neurons that nurture adds to nature. This personalisation of your brain, to borrow a phrase from the neuroscientist Susan Greenfield, is what Seung calls your “connectome”. Just as your genome identifies you as a unique biological animal (identical twins excepted), so, in theory, would your connectome – the complete mapping of the connections within your brain – identify you as a unique conscious individual.
That “in theory” turns out to be an enormous caveat. To analyse just one cubic millimetre of brain would yield one petabyte of information. Given that the entire human brain is a million times larger than this, we’re talking about the same amount of information as there is in a quadrillion (1,000,000bn) digital photographs. Seung is optimistic that the rapid development of computing technology will make this possible, and even if it doesn’t, mapping at a lower resolution would give us much clearer pictures of connectomes than is currently possible.
Seung believes that by mapping connectomes we could effectively become mind-readers. The problem with this, however, can be illustrated by the notorious Jennifer Aniston neuron. Experimenters found that although numerous neurons fired when a subject saw pictures or thought of women and actors, one only fired when thinking of Jennifer Aniston. There also turned out to be neurons just for Julia Roberts, Halle Berry, Kobe Bryant and so on.
Of course, that does not mean that each of these single neurons contained the subject’s entire idea of a particular actor. What such research suggests is that there is a unique pattern of neuronal firing for every distinct thought, memory or idea that we have. If that’s right, it would seem that, if we could learn to read the patterns, we could effectively see the content of people’s minds. Some have even suggested that by uploading these patterns on to computers we could effectively escape our biological wetware and run on electrical hardware, where biological ageing and physical limitations would no longer be a problem.
However, it’s not nearly as simple as that – if such feats of mind-reading and writing could ever be described as simple. First of all, it is certainly the case that the pattern of neuronal firing which produces the thought of Jennifer Aniston in you is not the same as that which produces it in me. So matching thoughts to neuronal patterns is a bit like matching mobile phones to their serial numbers. Once you know the serial number of a phone, you can identify it by its number. But that’s not because there is something about the number itself that enables you to work out it must refer to a particular phone. Indeed, there’s nothing about a serial number that tells you it refers to a phone at all. Likewise, knowing a neuronal pattern by itself wouldn’t reveal the thought it produces. No amount of looking at a connectome can reveal what its patterns mean unless you already know the content of the person’s experience.
What’s worse, given that our ideas of people and places change over time, it’s almost certain that the neuronal patterns that encode them change too. So what was the “Jennifer Aniston” neuron for that experimental subject in 2001 could well be her “Jubilee Victoria Sponge” neuron now. And all this assumes that connections alone are enough to determine the thought when, for all we now know, what’s going on inside neurons matters as much as, or even more than, what’s going on between them.
Trying to read off thoughts and experiences from brains is therefore something like trying to translate a language that only has one speaker and which is changing its pronunciation, spelling and lexicon every day.
To his credit, Seung admits that connectomics is largely in the business of proposing hypotheses that are as yet unproven. Nevertheless, he is surely right that there is value in pushing the project to see how far it can go. Although it is implausible that each brain would encode particular information in exactly the same way, perhaps there will turn out to be enough similarities for us to be able to know at least whether a person is thinking of a woman or a worm, a symphony or a sonnet.
More importantly, it does seem likely, since thoughts and emotional responses are the result of particular patterns of neuronal firing, that some psychological disorders and cognitive impairments are “connectopathies”: problems caused by the breakdown of healthy forms of neuronal connection. So, for example, we might be able to understand and treat dementia better if we find out exactly what types of neuronal connection it disrupts.
Part of Seung’s optimism that we’re going to get much better at reading connectomes is based on the fact that, until recently, the only tools we had to do so were crude, slow and expensive. As he rightly points out, many scientific advances arise only when the technology makes then possible. Adapting a saying by Francis Bacon, Seung coins the dictum: “Worthwhile things that have never yet been done can only be done by means that have never yet existed.”
Seung doesn’t know what we’ll find as we develop the means to map connectomes in ever more detail, but he is surely right to be excited at the new horizons of human knowledge that such breakthroughs will open up.
Julian Baggini’ is author (with Antonia Macaro) of ‘The Shrink and The Sage’ (Icon)
Please don't cut articles from FT.com and redistribute by email or post to the web.