Nervous system - Scholarpedia (original) (raw)

This article has not yet been published; it may contain inaccuracies, unapproved changes, or be unfinished.

The nervous system is the part of an animal's body that coordinates its behavior and transmits signals between different body areas. In vertebrates it consists of two main parts, called the central nervous system (CNS) and the peripheral nervous system (PNS). The CNS contains the brain and spinal cord. The PNS consists mainly of nerves, which are long fibers that connect the CNS to every other part of the body, but also includes other components such as peripheral ganglia, sympathetic and parasympathetic ganglia, and the enteric nervous system, a semi-independent part of the nervous system whose function is to control the gastrointestinal system.

At the cellular level, the nervous system is defined by the presence of a special type of cell, called the neuron, also known as a "nerve cell". Neurons have special properties that allow them to send signals rapidly and precisely to other cells. They send these signals in the form of electrochemical waves traveling along thin fibers called axons, which cause chemicals called neurotransmitters to be released at junctions to other neurons, called synapses. A cell that receives a synaptic signal from a neuron (a postsynaptic neuron) may be excited, inhibited, or otherwise modulated. The connections between neurons form neural circuits that can generate very complex patterns of dynamical activity. Along with neurons, the nervous system also contains other specialized cells called glial cells (or simply glia), which provide structural and metabolic support. Recent evidence suggests that glia may also have a substantial signaling role.

Nervous systems are found in almost all multicellular animals, but vary greatly in complexity. The only multicellular animals that have no nervous system at all are sponges and microscopic bloblike organisms called placozoans and mesozoans. The nervous systems of ctenophores (comb jellies) and cnidarians (e.g., anemones, hydras, corals and jellyfishes) consist of a diffuse nerve net. All other types of animals, with the exception of echinoderms and a few types of worms, have a nervous system containing a brain, a central cord (or two cords running in parallel), and nerves radiating from the brain and central cord. The size of the nervous system ranges from a few hundred cells in the simplest worms, to on the order of 100 billion cells in humans.

At the most basic level, the function of the nervous system is to control movement of the organism and to affect the environment (e.g., through pheromones). This is achieved by sending signals from one cell to others, or from one part of the body to others. The output from the nervous system derives from signals that travel to muscle cells, causing muscles to be activated, and from signals that travel to endocrine cells, causing hormones to be released into the bloodstream or other internal fluids. The input to the nervous system derives from sensory cells of widely varying types, which transmute physical modalities such as light and sound into neural activity. Internally, the nervous system contains complex webs of connections between nerve cells that allow it to generate patterns of activity that depend only partly on sensory input. The nervous system is also capable of storing information over time, by dynamically modifying the strength of connections between neurons, as well as other mechanisms.

Contents

Structure

The nervous system derives its name from nerves, which are cylindrical bundles of fibers that emanate from the brain and central cord, and branch repeatedly to innervate every part of the body. Nerves are large enough to have been recognized by the ancient Egyptians, Greeks, and Romans (Finger, 2001, chapter 1), but their internal structure was not understood until it became possible to examine them using a microscope. A microscopic examination shows that nerves consist primarily of the axons of neurons, along with a variety of membranes that wrap around them. The neurons that give rise to nerves do not generally lie within the nerves themselves — their cell bodies reside within the brain, central cord, or peripheral ganglia.

All animals more derived than sponges have nervous systems. However, even sponges, unicellular animals, and non-animals such as slime molds have cell-to-cell signalling mechanisms that are precursors to those of neurons (Sakarya et al., 2007). In radially symmetric animals such as the jellyfish and hydra, the nervous system consists of a diffuse network of isolated cells. In bilaterian animals, which make up the great majority of existing species, the nervous system has a common structure that originated early in the Cambrian period, over 500 million years ago.

Cells

The nervous system contains two main categories or types of cells: neurons and glial cells.

Neurons

The nervous system is defined by the presence of a special type of cell, the neuron (sometimes called "neurone" or "nerve cell"). Neurons can be distinguished from other cells in a number of ways, but their most fundamental property is that they communicate with other cells via synapses, which are junctions containing molecular machinery that allows rapid transmission of signals, either electrical or chemical. Many types of neuron possess an axon, a protoplasmic protrusion that can extend to distant parts of the body and make thousands of synaptic contacts. Axons frequently travel through the body in bundles called nerves (in the PNS) or tracts (in the CNS).

Even in the nervous system of a single species such as humans, hundreds of different types of neurons exist, with a wide variety of morphologies and functions. These include sensory neurons that transmute physical stimuli such as light and sound into neural signals, and motor neurons that transmute neural signals into activation of muscles or glands. In many species, though, the majority of neurons receive all of their input from other neurons and send their output to other neurons.

Glial cells

Glial cells (named from the Greek word for "glue") are non-neuronal cells that provide support and nutrition, maintain homeostasis, form myelin, and participate in signal transmission in the nervous system (Allen, 2009). In the human brain, it is currently estimated that the total number of glia roughly equals the number of neurons, although the proportions vary in different brain areas (Azevedo et al., 2009). Among the most important functions of glial cells are to support neurons and hold them in place; to supply nutrients to neurons; to insulate neurons electrically; to destroy pathogens and remove dead neurons; and to provide guidance cues directing the axons of neurons to their targets. A very important set of glial cell (oligodendrocytes in the vertebrate CNS, and Schwann cells in the PNS) generate layers of a fatty substance called myelin that wrap around axons and provide electrical insulation that allows them to transmit signals much more rapidly and efficiently.

Anatomy in vertebrates

Figure 1: Major divisions of the vertebrate nervous system.

The nervous system of vertebrate animals is divided into two parts called the central nervous system (CNS) and peripheral nervous system (PNS).

The CNS is the largest part, and includes the brain and spinal cord. The CNS is enclosed and protected by meninges, a three-layered system of membranes, including a tough, leathery outer layer called the dura mater. The brain is also protected by the skull, and the spinal cord by the vertebral bones. Blood vessels that enter the CNS are surrounded by cells that form a tight chemical seal called the blood-brain barrier, preventing many types of chemicals present in the body from gaining entry to the CNS.

The peripheral nervous system (PNS) is a collective term for the nervous system structures that do not lie within the CNS. The large majority of the axon bundles called nerves are considered to belong to the PNS, even when the cell bodies of the neurons to which they belong reside within the brain or spinal cord. The PNS is divided into "somatic" and "visceral" parts. The somatic part consists of the nerves that innervate the skin, joints, and muscles. The cell bodies of somatic sensory neurons lie in dorsal root ganglion of the spinal cord. The visceral part, also known as the autonomic nervous system, contains neurons that innervate the internal organs, blood vessels, and glands. The autonomic nervous system itself consists of two parts: the sympathetic nervous system and the parasympathetic nervous system. Some authors also include sensory neurons whose cell bodies lie in the periphery (for senses such as hearing) as part of the PNS; others, however, omit them (Hubbard, 1974, p. vii).

The vertebrate nervous system can also be divided into areas called gray matter ("grey matter" in British spelling) and white matter. Gray matter (which is only gray in preserved tissue, and is better described as pink or light brown in living tissue) contains a high proportion of cell bodies of neurons. White matter is composed mainly of myelin-coated axons, and takes its color from the myelin. White matter includes all of the body's nerves, and much of the interior of the brain and spinal cord. Gray matter is found in clusters of neurons in the brain and spinal cord, and in cortical layers that line their surfaces. There is an anatomical convention that a cluster of neurons in the brain is called a "nucleus", whereas a cluster of neurons in the periphery is called a "ganglion". There are, however, a few exceptions to this rule, notably the part of the brain called the basal ganglia.

Comparative anatomy and evolution

Neural precursors in sponges

Sponges have no cells connected to each other by synaptic junctions, that is, no neurons, and therefore no nervous system. They do, however, have homologs of many genes that play key roles in synaptic function in other animals. Recent studies have shown that sponge cells express a group of proteins that cluster together to form a structure resembling a postsynaptic density (the signal-receiving part of a synapse) (Sakarya, 2007). However, the function of that structure is currently unclear. Although sponge cells do not show synaptic transmission, they do communicate with each other via calcium waves and other impulses, which mediate some simple actions such as whole-body contraction (Jacobs et al., 2007).

Radiata

Jellyfish, comb jellies, and related animals have diffuse nerve nets rather than a central nervous system. In most jellyfish the nerve net is spread more or less evenly across the body; in comb jellies it is concentrated near the mouth. The nerve nets consist of sensory neurons, which pick up chemical, tactile, and visual signals; motor neurons, which can activate contractions of the body wall; and intermediate neurons, which detect patterns of activity in the sensory neurons and, in response, send signals to groups of motor neurons. In some cases groups of intermediate neurons are clustered into discrete ganglia (Ruppert et al., 2004).

The development of the nervous system in radiata is relatively unstructured. Unlike bilaterians, radiata only have two primordial cell layers, the endoderm and ectoderm. Neurons are generated from a special set of ectodermal precursor cells, which also serve as precursors for every other ectodermal cell type (Sanes et al., 2006).

Bilateria

Figure 2: Nervous system of a generic bilaterian animal, in the form of a nerve cord with segmental enlargements, and a "brain" at the front. (Note: this cartoon shows the nerve cord on the dorsal side of the body, but as the article explains, in protostomes it generally lies on the ventral side.)

The vast majority of existing animals are bilaterians, meaning animals with left and right sides that are approximate mirror images of each other. All bilateria are thought to have descended from a common wormlike ancestor that appeared during the Cambrian period, 550–600 million years ago (Balavoine, 2003). The fundamental bilaterian body form is a tube with a hollow gut cavity running from mouth to anus, and a nerve cord (or two parallel nerve cords), with an enlargement (a "ganglion") for each body segment, with an especially large ganglion at the front, called the "brain". It has not been definitively established whether the generic form of the bilaterian central nervous system is inherited from the so-called "Urbilaterian" — the last common ancestor of all existing bilaterians — or whether separate lines have evolved similar structures in parallel (Northcutt, 2012). On one hand, the presence of a shared set of genetic markers, as well as a tripartite brain structure shared by widely separated species (Hirth, 2010), suggest common derivation; on the other hand, the fact that some modern types of bilaterians (such as echinoderms) lack a central nerve cord, while many lack recognizably tripartite brains, suggest that this might have been the primitive state (Northcutt, 2012).

Vertebrates, annelids, crustaceans, and insects all show the segmented bilaterian body plan at the level of the nervous system. In mammals, the spinal cord contains a series of segmental ganglia, each giving rise to motor and sensory nerves that innervate a portion of the body surface and underlying musculature. On the limbs, the layout of the innervation pattern is complex, but on the trunk it gives rise to a series of narrow bands. The top three segments belong to the brain, giving rise to the forebrain, midbrain, and hindbrain (Ghysen, 2003).

Bilaterians can be divided, based on events that occur very early in embryonic development, into two groups (superphyla) called protostomes and deuterostomes (Erwin et al., 2002). Deuterostomes include vertebrates as well as echinoderms, hemichordates (mainly acorn worms), and Xenoturbellidans (Bourlat et al., 2006). Protostomes, the more diverse group, include arthropods, molluscs, and numerous types of worms. There is a basic difference between the two groups in the placement of the nervous system within the body: protostomes possess a nerve cord on the ventral (usually bottom) side of the body, whereas in deuterostomes the nerve cord is on the dorsal (usually top) side. In fact, numerous aspects of the body are inverted between the two groups, including the expression patterns of several genes that show dorsal-to-ventral gradients. Most anatomists now consider that the bodies of protostomes and deuterostomes are "flipped over" with respect to each other, a hypothesis that was first proposed by Geoffroy Saint-Hilaire for insects in comparison to vertebrates. Thus insects, for example, have nerve cords that run along the ventral midline of the body, while all vertebrates have spinal cords that run along the dorsal midline (Lichtneckert and Reichert, 2005).

Annelids

Figure 3: Earthworm nervous system. Top: side view of the front of the worm. Bottom: nervous system in isolation, viewed from above

Worms are the simplest bilaterian animals, and reveal the basic structure of the bilaterian nervous system in the most straightforward way. As an example, earthworms have dual nerve cords running along the length of the body and merging at the tail and the mouth. These nerve cords are connected to each other by transverse nerves resembling the rungs of a ladder. These transverse nerves help coordinate movement of the two sides of the animal. Two ganglia at the head end function as a simple brain. Photoreceptors in the animal's eyespots provide sensory information on light and dark (Adey, WR).

Ecdysozoa

Ecdysozoa are animals that shed their cuticle. These include nematodoes and arthropods.

Nematodes

The nervous system of one particular type of nematode, the tiny roundworm Caenorhabditis elegans, has been mapped out down to the synaptic level. This has been possible because in this species, every individual worm (ignoring mutations and sex differences) has an identical set of neurons, with the same locations and chemical features, and the same connections to other cells. Every neuron and its cellular lineage has been recorded and most, if not all, of the neural connections are mapped. The nervous system of C. elegans is sexually dimorphic; the nervous systems of the two sexes, males and hermaphrodites, have different numbers of neurons and groups of neurons that perform sex-specific functions. Males have exactly 383 neurons, while hermaphrodites have exactly 302 neurons (Hobert, 2005), an unusual feature called eutely.

Arthropods

Arthropods, such as insects and crustaceans, have a nervous system made up of a series of ganglia, connected by a pair of ventral nerve cords running along the length of the abdomen (Chapman, 1998). Most body segments have one ganglion on each side, but some are fused to form the brain and other large ganglia. The head segment contains the brain, also known as the supraesophageal ganglion. In the insect nervous system, the brain is anatomically divided into the protocerebrum, deutocerebrum, and tritocerebrum. Immediately behind the brain is the subesophageal ganglion, which is composed of three pairs of fused ganglia. It controls the mouthparts, the salivary glands and certain muscles. Many arthropods have well-developed sensory organs, including compound eyes for vision and antennae for olfaction and pheromone sensation. The sensory information from these organs is processed by the brain.

In arthropods, most neurons have cell bodies that are positioned at the edge of the brain and are electrically passive — the cell bodies serve only to provide metabolic support and do not participate in signalling. A protoplasmic fiber, called the primary neurite, runs from the cell body and branches profusely, with some parts transmitting signals and other parts receiving signals. Thus, most parts of the insect brain have passive cell bodies arranged around the periphery, while the neural signal processing takes place in a tangle of protoplasmic fibers called "neuropil", in the interior (Chapman, 1998). There are, however, important exceptions to this rule, including the mushroom bodies, which play a central role in learning and memory.

"Identified" neurons

A neuron is called identified if it has properties that distinguish it from every other neuron in the same animal — such as location, neurotransmitter, gene expression pattern, and connectivity — and if every individual organism belonging to the same species has one and only one neuron with the same set of properties (Hoyle and Wiersma, 1977). In vertebrate nervous systems very few neurons are "identified" in this sense — in humans, there are believed to be none — but in simpler nervous systems, some or all neurons may be thus unique. As mentioned above, in the roundworm Caenorhabditis Elegans every neuron in the body is uniquely identifiable, with the same location and the same connections in every individual worm.

The brains of many molluscs and insects also contain substantial numbers of identified neurons (Hoyle and Wiersma, 1977). In vertebrates, the best known identified neurons are the gigantic Mauthner cells of fish (Stein, 1999). Every fish has two Mauthner cells, located in the bottom part of the brainstem, one on the left side and one on the right. Each Mauthner cell has an axon that crosses over, innervating neurons at the same brain level and then traveling down through the spinal cord, making numerous connections as it goes. The synapses generated by a Mauthner cell are so powerful that a single action potential gives rise to a major behavioral response: within milliseconds the fish curves its body into a C-shape, then straightens, thereby propelling itself rapidly forward. Functionally this is a fast escape response, triggered most easily by a strong sound wave or pressure wave impinging on the lateral line organ of the fish. Mauthner cells are not the only identified neurons in fish — there are about 20 more types, including pairs of "Mauthner cell analogs" in each spinal segmental nucleus. Although a Mauthner cell is capable of bringing about an escape response all by itself, in the context of ordinary behavior other types of cells usually contribute to shaping the amplitude and direction of the response.

Mauthner cells have been described as "command neurons". A command neuron is a special type of identified neuron, defined as a neuron that is capable of driving a specific behavior individually (Stein, 1999, p. 112). Such neurons appear most commonly in the fast escape systems of various species — the squid giant axon and squid giant synapse, used for pioneering experiments in neurophysiology because of their enormous size, both participate in the fast escape circuit of the squid. The concept of a command neuron has, however, become controversial, because of studies showing that some neurons that initially appeared to fit the description were really only capable of evoking a response in a limited set of circumstances (Simmons and Young, 1999).

Function

The ultimate function of the nervous system is to control the body, especially its movement in the environment. It does this by extracting information from the environment using sensory receptors, sending signals that encode this information into the central nervous system, processing the information to determine an appropriate response, and sending output signals to muscles or glands to activate the response. The evolution of a complex nervous system has made it possible for various animal species to have advanced perceptual capabilities such as vision, complex social interactions, rapid coordination of organ systems, and integrated processing of concurrent signals. In humans, the sophistication of the nervous system makes it possible to have language, abstract representation of concepts, transmission of culture, and many other features of human society that would not exist without the human brain.

At the most basic level, the nervous system sends signals from one cell to others, or from one part of the body to others. There are multiple ways that a cell can send signals to other cells. One is by releasing chemicals called hormones into the internal circulation, so that they can diffuse to distant sites. In contrast to this "broadcast" mode of signaling, the nervous system provides "point-to-point" signals — neurons project their axons to specific target areas and make synaptic connections with specific target cells. Thus, neural signaling is capable of a much higher level of specificity than hormonal signaling. It is also much faster: the fastest nerve signals travel at speeds that exceed 100 meters per second.

Neurons and synapses

Figure 4: Major elements in synaptic transmission. An electrochemical wave called an action potential travels along the axon of a neuron. When the wave reaches a synapse, it provokes release of neurotransmitter molecules, which bind to chemical receptor molecules located in the membrane of the target cell.

Most neurons send signals via their axons, although some types are capable of emitting signals from their dendrites. In fact, some types of neurons such as the amacrine cells of the retina have no axon, and communicate only via their dendrites. Neural signals propagate along an axon in the form of electrochemical waves called action potentials, which emit cell-to-cell signals at points of contact called "synapses".

Synapses may be electrical or chemical. Electrical synapses pass ions directly between neurons (Hormuzdi et al., 2004), but chemical synapses are much more common, and much more diverse in function. At a chemical synapse, the cell that sends signals is called presynaptic, and the cell that receives signals is called postsynaptic. Both the presynaptic and postsynaptic regions of contact are full of molecular machinery that carries out the signalling process. The presynaptic area contains large numbers of tiny spherical vessels called synaptic vesicles, packed with neurotransmitter chemicals. When calcium enters the presynaptic terminal through voltage-gated calcium channels, an arrays of molecules embedded in the membrane are activated, and cause the contents of some vesicles to be released into the narrow space between the presynaptic and postsynaptic membranes, called the synaptic cleft. The neurotransmitter then binds to chemical receptors embedded in the postsynaptic membrane, causing them to enter an activated state. Depending on the type of receptor, the effect on the postsynaptic cell may be excitatory, inhibitory, or modulatory in more complex ways. For example, release of the neurotransmitter acetylcholine at a synaptic contact between a motor neuron and a muscle cell depolarizes the muscle cell and starts a series of events, which results in a contraction of the muscle cell. The entire synaptic transmission process takes only a fraction of a millisecond, although the effects on the postsynaptic cell may last much longer (even indefinitely, in cases where the synaptic signal leads to the formation of a memory trace).

There are literally hundreds of different types of synapses, even within a single species. In fact, there are over a hundred known neurotransmitter chemicals, and many of them activate multiple types of receptors. Many synapses use more than one neurotransmitter — a common arrangement is for a synapse to use one fast-acting small-molecule neurotransmitter such as glutamate or GABA, along with one or more peptide neurotransmitters that play slower-acting modulatory roles. Neuroscientists generally divide receptors into two broad groups: ligand-gated ion channels and G-protein coupled receptors (GPCRs) that rely on second messenger signaling. When a ligand-gated ion channel is activated, it opens a channel that allow specific types of ions to flow across the membrane. Depending on the type of ion, the effect on the target cell may be excitatory or inhibitory by bringing the membrane potential closer or farther from threshold for triggering an action potential. When a GPCR is activated, it starts a cascade of molecular interactions inside the target cell, which may ultimately produce a wide variety of complex effects, such as increasing or decreasing the sensitivity of the cell to stimuli, or even altering gene transcription.

According to Dale's principle, which has only a few known exceptions, a neuron releases the same neurotransmitters at all of its synapses (Strata and Harvey, 1999). This does not mean, though, that a neuron exerts the same effect on all of its targets, because the effect of a synapse depends not on the neurotransmitter, but on the receptors that it activates. Because different targets can (and frequently do) use different types of receptors, it is possible for a neuron to have excitatory effects on one set of target cells, inhibitory effects on others, and complex modulatory effects on others still. Nevertheless, it happens that the two most widely used neurotransmitters, glutamate and gamma-Aminobutyric acid (GABA), each have largely consistent effects. Glutamate has several widely occurring types of receptors, but all of them are excitatory or modulatory. Similarly, GABA has several widely occurring receptor types, but all of them are inhibitory. (There are a few exceptional situations in which GABA has been found to have excitatory effects, mainly during early development. For a review see Marty and Llano, 2005.) Because of this consistency, glutamatergic cells are frequently referred to as "excitatory neurons", and GABAergic cells as "inhibitory neurons". Strictly speaking this is an abuse of terminology — it is the receptors that are excitatory and inhibitory, not the neurons — but it is commonly seen even in scholarly publications.

One very important subset of synapses are capable of forming memory traces by means of long-lasting activity-dependent changes in synaptic strength. The best-understood form of neural memory is a process called long-term potentiation (abbreviated LTP), which operates at synapses that use the neurotransmitter glutamate acting on a special type of receptor known as the NMDA receptor (Cooke and Bliss, 2006). The NMDA receptor has an "associative" property: if the two cells involved in the synapse are both activated at approximately the same time, a channel opens that permits calcium to flow into the target cell (Bliss and Collingridge, 1993). The calcium entry initiates a second messenger cascade that ultimately leads to an increase in the number of glutamate receptors in the target cell, thereby increasing the effective strength of the synapse. This change in strength can last for weeks or longer. Since the discovery of LTP in 1973, many other types of synaptic memory traces have been found, involving increases or decreases in synaptic strength that are induced by varying conditions, and last for variable periods of time (Cooke and Bliss, 2006). Reward learning, for example, depends on a variant form of LTP that is conditioned on an extra input coming from a reward-signalling pathway that uses dopamine as neurotransmitter (Kauer and Malenka, 2007). All these forms of synaptic modifiability, taken collectively, give rise to neural plasticity, that is, to a capability for the nervous system to adapt itself to variations in the environment.

Neural circuits and systems

The basic neuronal function of sending signals to other cells includes a capability for neurons to exchange signals with each other. Networks formed by interconnected groups of neurons are capable of a wide variety of functions, including feature detection, pattern generation, and timing (Dayan and Abbott, 2005). In fact, it is difficult to assign limits to the types of information processing that can be carried out by neural networks: Warren McCulloch and Walter Pitts proved in 1943 that even artificial neural networks formed from a greatly simplified mathematical abstraction of a neuron are capable of universal computation. Given that individual neurons can generate complex temporal patterns of activity independently, the range of capabilities possible for even small groups of neurons are beyond current understanding.

Figure 5: Illustration of pain pathway, from René Descartes's Treatise of Man

Historically, for many years the predominant view of the function of the nervous system was as a stimulus-response associator (Sherrington, 1906). In this conception, neural processing begins with stimuli that activate sensory neurons, producing signals that propagate through chains of connections in the spinal cord and brain, giving rise eventually to activation of motor neurons and thereby to muscle contraction, i.e., to overt responses. Descartes believed that all of the behaviors of animals, and most of the behaviors of humans, could be explained in terms of stimulus-response circuits, although he also believed that higher cognitive functions such as language were not capable of being explained mechanistically. Charles Sherrington, in his influential 1906 book The Integrative Action of the Nervous System, developed the concept of stimulus-response mechanisms in much more detail, and Behaviorism, the school of thought that dominated Psychology through the middle of the 20th century, attempted to explain every aspect of human behavior in stimulus-response terms (Baum, 2005).

However, experimental studies of electrophysiology, beginning in the early 20th century and reaching high productivity by the 1940s, showed that the nervous system contains many mechanisms for generating patterns of activity intrinsically, without requiring an external stimulus (Piccolino, 2002). Neurons were found to be capable of producing regular sequences of action potentials, or sequences of bursts, even in complete isolation. When intrinsically active neurons are connected to each other in complex circuits, the possibilities for generating intricate temporal patterns become far more extensive. A modern conception views the function of the nervous system partly in terms of stimulus-response chains, and partly in terms of intrinsically generated activity patterns — both types of activity interact with each other to generate the full repertoire of behavior.

Reflexes and other stimulus-response circuits

Figure 6: Simplified schema of basic nervous system function: signals are picked up by sensory receptors and sent to the spinal cord and brain, where processing occurs that results in signals sent back to the spinal cord and then out to motor neurons

The simplest type of neural circuit is a reflex arc, which begins with a sensory input and ends with a motor output, passing through a sequence of neurons in between. For example, consider the "withdrawal reflex" causing the hand to jerk back after a hot stove is touched. The circuit begins with sensory receptors in the skin that are activated by harmful levels of heat: a special type of molecular structure embedded in the membrane causes heat to change the electrical field across the membrane. If the change in electrical potential is large enough, it evokes an action potential, which is transmitted along the axon of the receptor cell, into the spinal cord. There the axon makes excitatory synaptic contacts with other cells, some of which project (send axonal output) to the same region of the spinal cord, others projecting into the brain. One target is a set of spinal interneurons that project to motor neurons controlling the arm muscles. The interneurons excite the motor neurons, and if the excitation is strong enough, some of the motor neurons generate action potentials, which travel down their axons to the point where they make excitatory synaptic contacts with muscle cells. The excitatory signals induce contraction of the muscle cells, which causes the joint angles in the arm to change, pulling the arm away.

In reality, this straightforward schema is subject to numerous complications. Although for the simplest reflexes there are short neural paths from sensory neuron to motor neuron, there are also other nearby neurons that participate in the circuit and modulate the response. Furthermore, there are projections from the brain to the spinal cord that are capable of enhancing or inhibiting the reflex.

Although the simplest reflexes may be mediated by circuits lying entirely within the spinal cord, more complex responses rely on signal processing in the brain. Consider, for example, what happens when an object in the periphery of the visual field moves, and a person looks toward it. The initial sensory response, in the retina of the eye, and the final motor response, in the oculomotor nuclei of the brain stem, are not all that different from those in a simple reflex, but the intermediate stages are completely different. Instead of a one or two step chain of processing, the visual signals pass through perhaps a dozen stages of integration, involving the thalamus, cerebral cortex, basal ganglia, superior colliculus, cerebellum, and several brainstem nuclei. These areas perform signal-processing functions that include feature detection, perceptual analysis, memory recall, decision-making, and motor planning.

Feature detection is the ability to extract biologically relevant information from combinations of sensory signals. In the visual system, for example, sensory receptors in the retina of the eye are only individually capable of detecting "dots of light" in the outside world. Second-level visual neurons receive input from groups of primary receptors, higher-level neurons receive input from groups of second-level neurons, and so on, forming a hierarchy of processing stages. At each stage, important information is extracted from the signal ensemble and unimportant information is discarded. By the end of the process, input signals representing "dots of light" have been transformed into a neural representation of objects in the surrounding world and their properties. The most sophisticated sensory processing occurs inside the brain, but complex feature extraction also takes place in the spinal cord and in peripheral sensory organs such as the retina.

Intrinsic pattern generation

Although stimulus-response mechanisms are the easiest to understand, the nervous system is also capable of controlling the body in ways that do not require an external stimulus, by means of internally generated patterns of activity. Because of the variety of voltage-sensitive ion channels that can be embedded in the membrane of a neuron, many types of neurons are capable, even in isolation, of generating rhythmic sequences of action potentials, or rhythmic alternations between high-rate bursting and quiescence. When neurons that are intrinsically rhythmic are connected to each other by excitatory or inhibitory synapses, the resulting networks are capable of a wide variety of dynamical behaviors, including attractor dynamics, periodicity, and even chaos. A network of neurons that uses its internal structure to generate spatiotemporally structured output, without requiring a correspondingly structured stimulus, is called a central pattern generator.

Internal pattern generation operates on a wide range of time scales, from milliseconds to hours or longer. One of the most important types of temporal pattern is circadian rhythmicity — that is, rhythmicity with a period of approximately 24 hours. All animals that have been studied show circadian fluctuations in neural activity, which control circadian alternations in behavior such as the sleep-wake cycle. Experimental studies dating from the 1990s have shown that circadian rhythms are generated by a "genetic clock" consisting of a special set of genes whose expression level rises and falls over the course of the day. Animals as diverse as insects and vertebrates share a similar genetic clock system. The circadian clock is influenced by light but continues to operate even when light levels are held constant and no other external time-of-day cues are available. The clock genes are expressed in many parts of the nervous system as well as many peripheral organs, but in mammals all of these "tissue clocks" are kept in synchrony by signals that emanate from a master timekeeper in a tiny part of the brain called the suprachiasmatic nucleus.

References