To date, every study that identified practice-related changes in brain structure located these changes in grey matter. Now, for the first time, a paper by Scholz and colleagues in the journal Nature Neuroscience has identified training-induced changes in white-matter architecture. This has important implications for how we understand neurodevelopment.

What are grey (gray) and white matter?
Grey matter is a thin layer (less than half a centimeter thick) that forms the outermost part of the brain. It derives its grey color from neurons and unmyelinated (uninsulated) axons. Grey matter is considered the part of the brain that accomplishes the major cognitive chores – attention, language, learning, memory, perception, and thinking. As such, it was not surprising to discover that when we learn our grey matter changes.

White matter is different. It consists mainly of myelinated axons and does not contain dendrites. It is most commonly regarded as the support structure to the cerebral cortex, and interconnects different brain regions. Hitherto, changes in white matter as a result of learning had never been observed.

What did the study show?
Scholz and colleagues used diffusion tensor imaging (DTI) to measure white matter in 48 individuals (DTI is a relatively new neuroimaging technique capable of penetrating deep into the brain). Half the individuals were placed into a training group for 6 weeks, where the learned to juggle. The other half did no training for the same six weeks. All the participants were scanned before and after training.

Unsurprisingly, the training group showed significant differences in gray-matter density. What will come as a surprise to many is the observation of increased white matter volume in the training group, specifically the area underlying the right posterior parietal sulcus. The control group showed no such differences.

Why is this important?
In most neuroscience textbooks, white matter is given scant consideration, especially when it comes to learning. Changes in dendrites – dendritic growth – generally hog the headlines as hallmarks of neurodevelopment. When we learn, dendritic branches (little tree-like protrusions that spurt out from our neurons) form new synapses and build the connections that help us remember. White matter has no dendrites, so clearly there is something else going on.

The authors suggest that electrical activity in axons (as the result of learning) may regulate myelination in axons. Similarly, gross changes in the axon – the diameter or packing density – may be the cause.

Whatever, the reason, the white matter is changing. This is very important and promises to elevate the status of white matter as a correlate of neurodevelopment. Moreover, the countless MRI and fMRI studies that associate grey matter changes with everything from Internet use to the acquisition of wealth may only be telling half the story.

MRI techniques are not able to penetrate white matter, and is is quite conceivable that the myriad of studies that focus on the cerebral cortex are missing something. Researchers and educators take note – white matter really does matter!!

What did the study show?

Psychosis is a disordered cognitive state that can include disorganized thoughts, delusions, or hallucinations. It is a common symptom of schizophrenia and has been linked to a number of brain areas, including the the dorsolateral prefrontal cortex (DLPFC) and the hippocampus. Schizophrenia is also strongly associated with a number of genes, and a recent genome-wide association study identified a single nucleotide polymorphism in ZNF804A  as particularly important. Now, for the first time, a German research team has confirmed a link between ZNF804A and these brain structures. The group compared 115 participants who were either risk-allele carriers or non-risk-allele carriers and found differences in how their brains connect. For risk-allele carriers, connections were reduced within the DLPFC and also between left and right DLPFC . Conversely, risk-allele carriers showed increased connectivity between the DLPFC and hippocampal areas.

The study, primarily based at the University of Heidelberg, Germany, focused on a single nucleotide polymorphism (SNP) in ZNF804A (rs1344706)

Schizophrenia is a devastating, highly heritable brain disorder of unknown etiology. Recently, the first common genetic variant associated on a genome-wide level with schizophrenia and possibly bipolar disorder was discovered in ZNF804A (rs1344706). We show, by using an imaging genetics approach, that healthy carriers of rs1344706 risk genotypes exhibit no changes in regional activity but pronounced gene dosage–dependent alterations in functional coupling (correlated activity) of dorsolateral prefrontal cortex (DLPFC) across hemispheres and with hippocampus, mirroring findings in patients, and abnormal coupling of amygdala. , show that rs1344706 or variation in linkage disequilibrium is functional in human brain, and validate the intermediate phenotype strategy in psychiatry.

The study is important for a number of reasons. Firstly, it highlights the importance of connectivity (or dysconnectivity) as a neurobiological marker of schizophrenia. Secondly, it establishes ZNF804A as functional in the human brain. Thirdly, it is an example of how genome-wide association studies can align with anatomical data – to quote the authors, it affirms that “the pathophysiology of overt disease” can “mirror candidate gene effects”.

Functional Genomics and Big Picture Science

This third point is important to how we view science as a whole. Scientific research has traditionally relied upon reductionism as the primary means of discovery. To understand the world, reductionism tells us, we must strip it down to its barest elements. The sequencing of the human genome represents the ultimate triumph of this principle — the dis-assembly of an enormously complex living thing into its three billion molecular constituents. While unraveling the genomic structure ushered in a new era for biology and medicine, it laid bare a new problem—that of genomic function. Now that we have disassembled the machine, we have to figure out how to put it back together again.

Scientists have, in many ways, been forced to abandon reductionism and to look instead at the bigger picture – to see how things fits together. Increasingly we see large multi-disciplinary groups, each of which holds a different piece of the picture. In the current example, we have neuroimaging experts, who look at the gross structure of the brain, collaborating with geneticists, who look at m0lecules, and psychiatrists who look at behavior. These collaborations are exciting, because they integrate a number of different perspectives. Ultimately, this represents a triumph for “big picture” science.