Tools that show real brain power


With the invention of more and more remarkable scientific instruments we are blowing away the barriers to understanding how our brains function, writes Alan Finkel.


The video of the "glass brain", above, is a an anatomically-realistic 3D brain visualisation depicting real-time activity from signals from a person wearing an EEG (electroencephalographic) cap.

Galileo tinkered with the telescope and found the moons of Jupiter. Leeuwenhoek’s lens revealed a new world of microbes in a drop of water. We’ve always relied on our scientific tools to extend the limits of our universe.

But when it comes to probing the universe’s most complex creation, our tools have been a little limited. The human brain has proven hard to fathom. For decades we’ve had tools to probe individual cells or scan entire brains. But as for observing how brain cells create circuits, and ultimately things like consciousness, we’ve come up short.

But all that is changing. A new generation of probes are revealing the brain as never before, and the pace should only accelerate with two new big brain projects announced last year by the US and European Union (see Cosmos issue 52 – The human brain project). The developments at both ends of the scale have been staggering. Researchers are not only probing deeper into the microscopic world, like Google Maps they are panning out to take in the big view.

As an electrical engineer I helped develop automated microscopes that probed cellular responses to potential drugs. We found ourselves hard up against what was known as the “Abbe diffraction limit”, named after the nineteenth century German physicist Ernst Karl Abbe. He told us in 1873 that even with a perfect lens there was a fundamental limit to resolving small objects: you could never resolve anything smaller than approximately half the wavelength of the illuminating light.

With visible light, this limited us to observing features that are 200 nanometres or bigger. That’s just fine for looking at whole brain cells that are several thousand nanometres across, but far too coarse a probe for the 40-nanometre neurotransmitter capsules filling the synapses that relay signals from one brain cell to another.

But in a leap as dramatic as that provided by Galileo and Leeuwenhoek, the Abbe diffraction limit has been blown away in the past decade. A number of brilliant techniques now use scanning optical fibres and probabilistic approaches to resolve down to 20 nanometres, 10 times better than the Abbe diffraction limit. Brain researchers are now able to see details as fine as the doughnut shape of the neurotransmitter capsules.

Scientists don’t need to slice the brain to see inside: they make it transparent.

In truth, this defiance of limits is a little unsettling. But I’ve seen it before. When I was a student, it was commonly accepted that a copper telephone line was limited to a four kilohertz bandwidth, which meant that using basic data modulation schemes they could carry 8,000 bits per second of data. Today, those same copper telephone lines have been tweaked to bring up to 20 million bits per second of data into our homes, an amazing speed up.

But I digress.

While it is illuminating to go microscopic, to truly understand the brain we also need to zoom out. MRI scanners do that and they’ve let us peer into the living brain for decades now. The problem is that they are low-resolution, good at showing us big structures like a blood clot or tumour that are medically invaluable, but not so good at displaying brain circuits. All that is about to change as well: advanced processing algorithms highlight the connections from one side of the brain to the other, and the best of the next generation of MRI scanners will resolve images at the 0.1 millimetre level, 10 times better than today's machines.

Even that resolution limit can be pushed, although not in a living brain. Researchers have been developing methods where the entire brain of an animal can be analysed down to the level of its subcellular components. Every tiny dendritic process that connects cells and represents the physical manifestation of learning and memory can be traced. Stephen Smith and colleagues at Stanford University, for instance, are developing a technique known as array tomography where brain tissue is sliced into sections just 70 nanometres thick. A computer then aligns and collates the images from each slice to build a high-resolution three-dimensional image of the brain tissue.

The latest advance, known as “Clarity”, from Karl Deisseroth and colleagues also at Stanford University, reveals much the same ultra-high resolution picture. The difference is they don’t need to slice the brain to see inside: they make it transparent.

These Google Map-style approaches are the most exciting. They offer the possibility of eventually understanding how brain circuitry connects up to generate humour, love or mathematical reasoning.

The pace of advance is truly mind-boggling, but we still have our work cut out for us. If there is one thing we’ve learned from a century of neuroscience research, it’s that the more we learn the more we realise how little we know about the complexities of the brain. As botanist Lyall Watson put it, “If the brain were so simple we could understand it, we would be so simple we couldn’t”.

Latest Stories
MoreMore Articles