Measure of the mind: connectome
Ransom Stephens- October 10, 2012Can a person’s thoughts be predicted from the configuration of their neurons?
H. Sebastian Seung is a Professor of Computational Neuroscience at MIT and the author of Connectome: How the brain’s wiring makes us who we are. A “connectome” is to neurons as the “genome” is to genes. It’s the map of every interconnection of every neuron.
Neurons are cells that behave much like multi-level logic gates, sort of semi-analog in that different neurons are triggered by different signal levels but once fired, neurons are essentially binary – or at least that’s the current state of the art. (Neuroscience, like most of biology, is so far from being a mathematical science that it’s safe to add “or at least that’s the current state of the art” to many such statements.)
Like all cells, neurons consist of a nucleus and cell body. What makes them special is that the cell body can reach across the brain and body with interconnects called axons and receivers called dendrites. The importance of neurons is in their connections. Synapses form at the connections and when a neuron fires, it generates small voltage at the synapse that can trigger another neuron which can trigger another. The sum of the interactions results in thoughts and actions from the desire for yet another cup of coffee, to cussing out your boss, to creating new technology.
Figure: Artist's concept of a Connectome (photo courtesy The National Institute of Health)
Seung believes that the connectome determines what thoughts are available to you. It’s not a static system, of course, the connections and firing thresholds evolve are altered by experience. His idea is that if you know the function of each neuron and their interconnections then you have everything you need to predict their reaction to any stimulus.
You might hope that you could map out someone’s connectome with some tidy test and measurement equipment but no such luck. The problem is that state of the art imaging technology, fMRI (functional magnetic resonance imaging), CT-scan (multi-angle x-rays), and PET-scan (positron emission tomography) have resolutions no better than millimeters and the wires that connect neurons to each other (that is, axons) are 10 microns or so in diameter.
Measuring a connectome is a messy business. The brain is sliced in 0.1 mm sheets and axons are mapped by tracking them among sheets. The only connectome that has so far been measured is that of a tiny roundworm, Caenorhabditis elegans, which has only 300 neurons with a few thousand connections scattered around its body – it doesn’t have a brain per se. The human brian, on the other hand, has about 86 billion neurons with an average of 10,000 connections each.
While measuring the connectome of a living subject is possible, whether or not it would provide the ultimate predictive insight into someone’s motivations and actions is up in the air. Should you accept Seung’s hypothesis that the physical configuration and individual functions of neurons form a complete description of a person’s mind poses an interesting philosophical question regarding the notion of free will.
Personally, I think that Seung’s hypothesis is true, but I don’t think that it is sufficient to contradict free will. A person’s connectome is derived from both their genetics and their experiences; that is, it’s built on the nonlinear feedback loop of a person’s choices. Nonlinear systems can be deterministic without being predictive.
(Of course, if Ransom has free will then his opinion is probably wrong.)