Thursday, 20 October 2016
Psychologist Scans Own Brain Twice per Week for a Year and a Half
If you use social media, you have probably seen posts where someone has shown photos of themselves taken at regular intervals over a long period. Well, Stanford University psychologist Russell Poldrack has gone them one better: he has scanned his own brain twice per week for a year and a half! But Poldrack’s goal isn’t simply to wow his friends on Facebook. He is using the scans to do something that has never been attempted before: to understand how the connectivity of a normal person’s brain may vary over a period of several months, a span of time in which people with mental disorders often show considerable fluctuations in their psychological functions.
Poldrack calls his study “MyConnnectome”, and he published his initial results in the December 9, 2015 edition of the journal Nature Communications. It might seem surprising that no one had ever gathered such data before. But not many normal subjects would have been willing to do what Poldrack did: get into an MRI machine for a brain scan two mornings each week (one of them on an empty stomach) for a year and a half, have blood samples taken once per week, and write a report on his diet and physical activity every day. It took a scientist who was really motivated to advance the state of knowledge.
This project has generated reams of data that researchers will continue to analyze for years to come. For example, the data from Poldrack’s blood samples will be used to correlate his brain’s connectivity with the physiological condition of his body as a whole. Sequencing of messenger RNA from his white corpuscles will allow the degree of expression of certain genes to be determined, so that correlations can then be drawn between gene activity and brain activity. And his reports on what he ate and what exercise he did on the day of any given scan will allow relationships to be established among diet, behaviour, gene expression, and brain activity.
Clearly, all this analysis will take some time. But certain relationships have already become fairly obvious. For example, Poldrack has found a strong correlation between his outbreaks of psoriasis (a genetic autoimmune skin disease that can be triggered by environmental factors) and the expression of genes related to inflammation and immune response. He has also found other correlations, between the state of his body and that of his brain, though he has not yet been able to explain them.
One of the most obvious and surprisingly strong correlations emerged from his having undergone his Tuesday-morning brain scans on an empty stomach, but his Thursday-morning scans after a good breakfast that included a cup of coffee. The psychostimulant effects of caffeine on the brain are well known, but it was not previously known just how much caffeine could rapidly influence the functional connectivity of the brain—in other words, which neural pathways the nerve impulse actually travels, out of all of the possible pathways (what is known as the anatomical connectome). The hypothesis here is that the parts of the brain that “talk” to one another preferentially thus form various brain networks that are subsets of all the possible associations offered by the anatomical connectome. These preferential pathways would be like the major highways between two big cities, as opposed to all of the other possible secondary routes between City A and City B.
On the mornings when Poldrack’s brain was steeped in caffeine, its connectivity was quite different from the mornings on which he had fasted. In particular, the connections between the somatosensory/motor network and the systems responsible for detailed analysis of visual stimuli were more intimate without caffeine. This observation was surprising not only because of the large size of the effect measured, but also because the areas of the brain affected were relatively “low-level” (as opposed to the more associative cortex responsible for “high-level” cognitive functions). For the moment, Poldrack does not want to overinterpret these findings, but he does speculate that the fatigue that he felt on the mornings that he had not had any caffeine may have caused his brain to allocate more resources to basic mechanisms such as visual and sensorimotor integration.
This project exemplifies two increasingly dominant approaches in cognitive science: connectome mapping and big-data analysis (the latter is also used in many other sciences, as the data-storage capabilities of modern information systems continue to grow). This project also tells us a lot about the current limitations of these two approaches.
As regards connectome mapping, we know that it is impossible to map the entire connectome of a brain definitively, because of the intrinsic plasticity of its synapses. In other words, the secondary pathways in any individual’s brain are constantly changing, and the best we can hope is to map its “main highways”. Poldrack’s MyConnectome project is interesting because it lets us accept this limitation by focusing chiefly on the typical functional connectivity of an individual’s brain over fairly long periods (because, of course, these networks are dynamic and are being torn down and rebuilt constantly).
As regards big data, the problem is the same whatever the application: classifying and analyzing the astronomical volumes of information involved. To facilitate this work, Poldrack and his team have decided to make all of their data available for free on the MyConnectome project website.
Sunday, 25 September 2016
How Different Parts of the Brain Co-operate
Perhaps one of the hardest things to understand about the brain is the way that it is organized into networks. In this post, I will discuss a 2015 study, on the brain structures involved in delayed gratification, that makes this complex subject a bit easier to grasp. (more…)
Wednesday, 7 September 2016
Motor cortex is required for learning but not for executing a motor skill
The motor cortex was long thought to be the part of the brain that controlled the body’s voluntary movements. Given the plasticity of the cortex as a whole, it seemed reasonable to believe that decisive changes in the connectivity of the neurons in the motor cortex might well be associated with motor learning. Although this may indeed be the case, a study published by Risa Kawai and colleagues in the journal Neuron in May 2015 forces us to reconsider the primacy of the motor cortex in learned sequences of movements, at least in rats. (more…)
Monday, 22 August 2016
A good general book on neuroscience
Whenever I give a presentation about the human brain, someone almost always comes up afterward and asks me whether I could recommend a good general book on neuroscience. In fact, there are several such books, but the one that I want to recommend here today offers a special advantage: you can buy a printed copy, but you can also access the entire book on the Internet for free!
The book that I’m talking about is Neuroscience, 2nd edition, edited by Dale Purves et al. and published by Sinauer Associates in 2001. It covers wide areas of modern neuroscience and is chock-full of very informative figures and tables. (more…)
Monday, 8 August 2016
A Nanometric 3D Representation of a Mouse Cortex Cortex
The analogy between a real forest and a “forest of neurons” has been drawn many times, but the images produced most recently by the team of Jeff Lichtman and Narayanan Kasthuri (see the first two links below) make it clear yet again that the complexity of the brain’s connections far surpasses that of the densest forest.
As Lichtman has long said, by going down to the scale of the electron microscope and then reconstructing the slightest contacts between axons, dendrites, and neighbouring glial cells slice by slice, we can detect patterns that escape us at the more “macro” scales previously used to model neuronal connectivity. (more…)