Connectivity—A Primer

by Carl Sherman

March 6, 2013

Neurons love company. The very basis of thought, feeling, and behavior is the passage of electrical signals from one to another; these circuits organize into increasingly complex structures, which may link up with comparable modules in distant regions of the brain.

In essence then, the human brain is a network of networks, an integrated system of stupefying complexity that somehow coordinates the operations of billions of units. From simple sensorimotor functions like the perception of a flash of light or directed movement of a finger, to complex cognitive operations like rational deliberation that balances memory, moral judgment, and imagination, mental life grows out of connectivity.

With a growing appreciation of its centrality to brain function, and the emergence of technology to illuminate its intricacies, connectivity has become a major focus of neuroscience research.

A brief history i, ii
Modern notions of connectivity have their roots in two opposing, centuries-old theories of the relation between brain structure and function.

Locationalism, which assigns behavioral and cognitive functions to sharply delimited parcels of the cerebral cortex, originated in the simplistic concepts of phrenology. According to the theory, elaborated by German physicians Franz Josef Gall and Johann Spurzheim, the strength of traits and capacities like love, curiosity, cautiousness, self-esteem, and hope are determined by the size of their corresponding cerebral segments, which may be deduced from palpable bumps on the skull.

The holistic or equipotentiality theory held that mental phenomena are generated by the whole brain functioning as a unified unit, and that attempts to equate function with region are intrinsically misleading.

The synthesis of these views began as a finer appreciation of the brain’s workings grew out of 19th century discoveries like Broca’s and Wernicke’s identification of language centers that are located in the left hemisphere but linked to other parts of the brain, and later delineations of the widely distributed systems involved in memory.

Some call today’s mind-brain model of neurons, circuits, and regions whose workings are at once place-specific and thoroughly interconnected “modified locationalism.”

The connectome i, iii, iv, v, vi 
This overarching vision of the brain as a multilayered web is best expressed in the pursuit of the connectome, which if fully realized would constitute a detailed map of the brain’s wiring in its entirety.

The term was coined in a 2005 paper by Olaf Sporns and colleagues, as an analogy to the genome, the sum total of genes that make up an organism. And like the genome, the concept has inspired a major, federally-funded research initiative, the Human Connectome Project, which collates and makes available to researchers accumulating data that add to our collective understanding of how the brain is organized.

A matter of scale
Connectivity in the brain is best understood on various levels of scale. Microscopically, it defines the fine-grained network of neuron-to-neuron connection via synapses. With billions of neurons, each connected to perhaps 10,000 others, complexity at this level is staggering.

On intermediate levels, assemblies or columns of connected neurons are organized into larger modules within a brain region that function together, such as the auditory cortex or structures like the hippocampus.

At the highest level, connections crisscross from region to region, linking modules and structures across (relatively) great distances within the brain.

While connectomics ultimately aims to map all connections from neuron to whole-brain scale, most attention, and progress, has involved higher-level structures.

Exploring connectivity i, ii, vii, viii, ix, x,xi
Investigating how the brain is organized, researchers use brain imaging and computer modeling to characterize three different, although related, types of connectivity.

Structural connectivity, the principal concern of connectomics, refers to the web of anatomical links that physically join parts of the brain to one another. On the neuron-to-neuron level, mapping synapses within the tangle of billions of cells and their projections obviously represents a staggering challenge, but advances in electron microscopy, the use of genetically coded fluorescent proteins to visualize neurons in hundreds of distinctive colors (“Brainbow”), and highly sophisticated computational systems point toward progress.

On a larger, more manageable scale, MRI techniques like diffusion tensor imaging (DTI) have made it possible to detail white matter pathways—bundles of axons that join parts of the brain the way fiberoptic cables transmit messages over long distances. These include commissural tracts like the corpus collosum, which connects the right and left hemisphere, and association tracts like the arcuate fasciculus, which links the frontal lobe with other parts of the cerebral cortex.

Functional connectivity characterizes how brain regions work together. Researchers infer this web of interactions by analyzing fMRI and EEG data reflecting activation in various parts of the brain, looking for patterns that emerge at rest or as subjects perform various tasks. Such studies suggest a fundamental architecture of “hubs” and “spokes”—relatively few highly connected areas serve a critical function in coordinating communication all across the brain—and illuminate the specifics of neural choreography behind operations as simple as identifying odors and as complex as moral judgment.

Effective connectivity goes a step further, elucidating how one part of the brain regulates activity in another. To deduce causal relationships between brain regions that appear to act together in generating thought, emotion, and behavior—which areas lead, and which follow— researchers use highly sophisticated computational methods, some derived from other disciplines like economic analysis.

A work in progress xii, xiii, xiv, xv, xvi, xvii
Connectivity changes over the lifespan, particularly during childhood and adolescence. Synaptic pruning thins out the dense underbrush of connections between neurons, sculpting circuits essential for normal function. At the same time, myelination—an insulating fatty sheath grows over axon fibers—bolsters communication within and between brain regions.

Recent DTI studies have provided insight into the neural circuitry of newborns, and suggested how differences in the rate at which white matter tracts develop could predict reading ability in school-aged children. Studies comparing children, adolescents, and adults have indicated that enhanced effective connectivity strengthens the ability of higher brain areas to control behavior with maturity, and suggest that more efficient white matter function is a key factor.

Certain neural circuits can only be shaped properly for a limited time: the cortical area that governs sight must be stimulated by appropriate visual experiences during the first few years of life for the eyes’ ability to focus to develop properly, for example. Similar critical periods circumscribe the fine-tuning of motor, language, and cognitive systems. When these periods end, physical phenomena like the laying down of myelin and shifts in the balance of neurotransmitters and other brain chemicals make it much harder for patterns of neural connection to change.

Research has clarified these braking mechanisms and suggested ways in which they can be overcome, at least partially, after the critical period. There is reason to believe drugs (e.g. certain antidepressants), physical exercise, and exposure to environmental stimuli in adulthood may restore a degree of neuroplasticity to even highly stable circuitry, such as the visual system.

Broadly speaking, small changes in structural connectivity are continual throughout life. Such basic mental phenomena as learning, memory, and the making and breaking of habits involve the strengthening and weakening of existing synapses and the creation of new ones.

Among other factors, corticosteroids have been shown to be essential for neuroplasticity in both developing and mature brains, while chronically excessive levels have a disruptive effect. (This might account for the link between stress, which is associated with a sustained increase in corticosteroid secretion, and psychiatric illness.)

The connected brain in sickness and health xviii, xix, xx, xxi, xxii, xxiii, xiv
Analyzing connectivity has yielded insights into the neural processes behind the experiences of everyday life.

It has added to our understanding of complex mental phenomena like the response to music (even suggesting how classical, country, and rock music influence the brain differently). Studies of structural and functional connectivity illuminate the coordinated neural events involved in reading, and help explain difficulties such as dyslexia.

There is substantial evidence that the pathological processes behind many brain disorders aren’t limited to one or another region but encompass far-flung networks. Autism very likely represents such a “connectopathy,” although the details are elusive. Some studies suggest that certain areas (such as those involved in speech) are underconnected, while on a broader scale overconnection is the problem.

Disrupted connectivity in schizophrenia has been an active research area, and recent studies have identified abnormalities, including interactions between language and auditory regions, that may help explain specific symptoms like auditory hallucinations and delusions.

In addition to the effects of neuron death at the synapse level, Alzheimer’s disease is known to disrupt both structural and functional connectivity of widespread brain networks.

With more precise brain mapping and a fuller understanding of the processes that govern connectivity may come better diagnosis and new treatments for these and other disorders.

i. Seung, Sebastian. Connectome. Harcourt Mifflin. 2012

ii. Filley, Christopher M. Neurobehavioral Anatomy, Third Edition. Boulder, CO: University Press of Colorado, 2011

iii. Valeo, T., Building a vision of the networks of the brain. Dana Foundation website; March 2, 2010.
iv. Sporns, O. (2009) Connectome. Scholarpedia, 5(2):5584

v. Sporns, O., Tononi, G. & Kotter, R. (2005) The human connectome: A structural description of the human brain. PLoS Comput Biol, 1, e42

vi. Human Connectivity Project 

vii. Leergaard, TB et al. Mapping the connectome: multi-level analysis of brain connectivity. Frontiers in Neuroinformatics, May 1, 2012

viii. Sieving, PA. A decade after The Decade of the Brain: using breakthroughs in visual neuroscience to treat diseases. Cerebrum, Feb 24, 2010

ix. Friston, K. Causal modelling and brain connectivity in functional magnetic resonance imaging. PLoS Biology; Feb 2009, Volume 7 (2): e33

x. Penny WD and Friston KJ (2007) Functional imaging. Scholarpedia, 2(5):1478

xi. Hwang K, Hallquist MN, Luna B.The development of hub architecture in the human functional brain network. Cereb Cortex 2012 Aug 28, 2012 [Epub ahead of print]

xii. Tymofiyeva O et al. Towards the ‘‘baby connectome’’: mapping the structural connectivity of the newborn brain. PLoS ONE, Feb 2012, Volume 7 (2):e31029

xiii. Yeatman JD et al. Development of white matter and reading skills. PNAS; Oct 30, 2012. Volume 109 no. 44: E3045–E3053

xiv. Hwang K, Velanova K, & Luna, B. Strengthening of top-down frontal cognitive control networks underlying the development of inhibitory control: a functional magnetic resonance imaging effective connectivity study. J. Neuroscience, Nov 17, 2010; 30(46):15535-15545

xv. Asato MR et al. White Matter development in adolescence: a DTI study. Cerebral Cortex, Sept 2010; 20:2122—2131

xvi. Liston C, Gan WB. Glucocorticoids are critical regulators of dendritic spine development and plasticity in vivo. Proc Natl Acad Sci U S A. 2011 Sept 20; 108(38):16074-9. Epub 2011 Sept 12.

xvii. Bavelier D, Levi DM et al. Removing brakes on adult brain plasticity: from molecular to behavioral interventions. J Neurosci. 2010 Nov 10; 30(45):14964-71.

xviii. Wilkins RW et al. Network Science: A new method for investigating the complexity of musical experiences in the brain. Leonardo, Volume 45, Number 3, 2012, pp. 282-283

xix. Hoeft F et al. Neural systems predicting long-term outcome in dyslexia. Proc Natl Acad Sci USA, Jan 4 2011; 108(1):361-6

xx. van der Mark S et al. The left occipitotemporal system in reading: disruption of focal fMRI connectivity to left inferior frontal and inferior parietal language areas in children with dyslexia. Neuroimage, Feb 1, 2011; 54(3):2426-36

xxi. Müller, R-A et al. Underconnected, but how? A survey of functional connectivity MRI studies in autism spectrum disorders. Cereb. Cortex, 2011; 21 (10): 2233-2243.

xxii. Karbasforoushan H, Woodward ND. Resting-state networks in schizophrenia. Current Topics in Medicinal Chemistry, Dec 21, 2012

xxiii. Fornito A et al. Schizophrenia, neuroimaging and connectomics. Neuroimage, Oct 1, 2012; 62(4):2296-314xiv. de Haan W et al. Disruption of functional brain networks in Alzheimer's disease: what can we learn from graph spectral analysis of resting-state magnetoencephalography? Brain Connect, 2012; 2(2):45-55