Neural Information Processing

Abstract

The central nervous system (CNS) is specialised in processing information originating from a variety of internal and external sources. Its basic information processing units are nerve cells, or neurons. Within the CNS, these elementary building blocks are densely interconnected in hierarchical and parallel pathways. Information originating from sensory neurons in contact with the body periphery is gradually transformed along these pathways to generate specific actions through signals relayed by motor neurons to peripheral organs. The information processing capabilities of animal brains are quite distinct from those of existing manā€made machines, being characterised by their resilience to noise, their capacity to learn and generalise, as well as their ability to participate in complex social behaviours.

Key Concepts:

  • Understanding neural information processing will require a detailed characterisation of how nerve cells code for sensory stimuli and motor commands, often referred to as the Neural Code.

  • Neural codes are predominantly based on spikes, which are discrete events in time and space allowing nerve cells to communicate with one another.

  • Neural codes most likely include spike rate codes and relative spike timing codes.

  • In many brain areas spikes are sparse across nerve cells and time.

  • The organisation of such neural codes is only loosely constrained in time and highly flexible.

Keywords: action potential; information processing; synapse; neuron; map

Figure 1.

Information processing by nervous systems. Sensory stimuli are transduced at the periphery by receptor cells and action potentials travel along the axons of sensory neurons towards the central nervous system (CNS). In the CNS, sensory information is processed by interneurons, eventually causing motor neurons to fire action potentials issuing specific motor commands for peripheral muscles. (a–d) illustrate in more detail some steps of this process. (a) In hair cells of the inner ear, transduction is caused by mechanical movement of cilia located at the tip of the receptors (double arrow). (b) This generates a receptor potential and the release of neurotransmitter on to the sensory neuron terminal, which in turn generates a postsynaptic potential causing an action potential to travel along the sensory axon. In the CNS, information is conveyed between neurons by chemical synapses. (c) An action potential causes the release of neurotransmitter that binds to receptors located on the postsynaptic cell membrane. This opens channels across the membrane allowing ions to flow and modify the cell's membrane potential. Synapses are usually located in the dendrites or on the cell body of neurons. (d) Synapses close to the spike initiation zone (the axon hillock) are often inhibitory; their activation causes the membrane potential to decrease with respect to its resting value, as illustrated by the inhibitory postsynaptic potential (IPSP) at synapse 3. When activated separately, synapses 1 and 2 generate excitatory postsynaptic potentials (EPSP 1 and 2). Often, simultaneous activation results in a change in membrane potential (dashed line of summation) that is smaller than the algebraic sum of the two individual EPSPs (EPSP 1+2, solid line of summation), indicative of nonlinear interactions between the two synaptic inputs. The summation of synaptic potentials is not strictly additive because it results from the opening of ion channels and the flow of ions across the cell membrane. This ion or current flow depends not only on the permeability or conductance of the open channels but also on the membrane potential itself, which changes dynamically as new synaptic events constantly impinge on the neuron. Motor neurons make contact with single muscle fibres at the neuromuscular junction. Sensory information also originates in muscles via several types of sensory afferent fibres such as muscle spindles.

Figure 2.

Common principles of CNS information processing. (a) In monkeys, visual information is processed by more than 30 densely interconnected subcortical nuclei and cortical areas (only a few areas and connections are illustrated here). The areas on the left (from V1 to VIP and MST), as well as the subcortical magnocellular pathway, process mainly spatial and motion information. They are located in the occipital and parietal lobes (see the upper arrow in the lateral view of the monkey brain on top). The areas on the right (from V1 to IT), as well as the parvocellular pathway, are involved in object recognition and are located in the occipital and temporal lobes (lower arrow on top). Abbreviations: posterior (p), anterior (a), ventral (v), dorsal (d), magnocellular pathway (M), parvocellular pathway (P), lateral geniculate nucleus of the thalamus (LGN), visual area 1 (V1), visual area 2 (V2), middle temporal area (MT), ventral posterior area (VP), ventral intraparietal area (VIP), medial superior temporal area (MST), inferotemporal area (IT). Adapted from Felleman and Van Essen (1991) Cerebral Cortex1: 1–47, with permission from Oxford University Press; Distler et al. (1993) Journal of Comparative Neurology334: 125–150. Copyright © 1993. Reprinted by permission of Wiley‐Liss, Inc., a subsidiary of John Wiley & Sons Ltd. (b) In the barn owl optic tectum (a brain structure equivalent to the mammalian superior colliculus) neurons are tuned to the location of sound sources in space. A reticular grid superimposed on the surface of the optic tectum indicates the location of sound sources eliciting optimal responses (maximal firing rates) for neurons at a particular location. The hatched area indicates the geometrical locus in space eliciting responses greater than 50% of maximum for a neuron centred close to 0 azimuth and elevation (this locus is often called the response field or receptive field of the cell). The top drawing illustrates how azimuth and elevation are defined (0 azimuth and elevation corresponds to a point in front of the animal in the median plane of the eyes). Adapted from Konishi (1986) Trends in Neurosciences9: 163–168. Copyright © 1986, with permission from Elsevier Science; Cohen and Knudsen (1999) Trends in Neurosciences22: 128–135. Copyright © 1999, with permission from Elsevier Science. (c) Lateral inhibition sharpens the response to stimulus edges. The top panel shows a stimulus with a sharp jump in mean value (such as a one‐dimensional bar that is dark on the left and bright on the right). The middle panel illustrates the receptive fields of the six receptor cells shown in the network diagram below. The response of each cell is proportional to the stimulus intensity and depends on the spatial position of the stimulus. The bottom panel illustrates the response profile of interneurons after processing by a lateral inhibitory network. The peak output activity is centred on the edge of the stimulus.

Figure 3.

Diversity of neuronal shape and function. (a) Example of three types of neurons found in the cerebellum, a hindbrain structure involved in the execution of complex motor programmes. Purkinje cells (top; bar, 100 μm) are inhibitory output elements, which send their axons (arrow) to the deep cerebellar nuclei and other regions of the brainstem. They have an extensive dendritic tree that arborises towards the surface of the cerebellum. A major source of excitatory input to Purkinje cell dendrites is relayed by granule cell axons, called parallel fibres. Granule cells (bottom left; bar, 10 μm) are tiny interneurons possessing only four dendrites, each receiving excitatory input (arrows) from extracerebellar brain regions. In addition, parallel fibres contact inhibitory interneurons called stellate cells (bottom right; bar, 100 μm) which also synapse on to different regions of the Purkinje cell dendritic tree (cells stained in the turtle). (b) Schematic summary of synaptic inputs to Purkinje cells (note that excitatory and inhibitory inputs shown on left and right are intermixed in the animal). In addition to the excitatory input mediated by parallel fibres and the inhibitory input from stellate cells, a second powerful source of excitatory input on proximal dendrites is provided by climbing fibres, which are the axons of neurons located in the inferior olive, a brainstem nucleus. (c) Gain control mechanism in the Omega neuron of the cricket. The bottom trace illustrates an experiment during which simulated male calling songs of strong (large pulses) and then weak intensity (small pulses) were presented while the membrane potential was recorded (top trace). Initially, the neuron responds with a vigorous burst of spikes to strong songs and by three spikes to a weak song (arrow). The response to weak songs rapidly decreases, reflecting an adjustment in the spiking threshold of the cell. This gain control mechanism is thought to be mediated by the influx of calcium into the cell, which can be monitored with fluorescent dyes and a camera system, as illustrated on top. Adapted from Sobel and Tank (1994) Science263: 823–826. Copyright © 1994. American Association for the Advancement of Science; Yuste and Tank (1996) Neuron16: 701–716. Copyright © 1996 Cell Press. (d) The dendritic tree of a giant tangential cell (VS8) in the brain of the fly is illustrated at the bottom left. This cell is sensitive to the motion of a small object over a large portion of the visual field. The direction of motion eliciting the strongest response varies systematically from location to location as illustrated on the right (larger arrows correspond to stronger responses; the arrow's direction indicates the motion direction eliciting the strongest response – see Figure b for a definition of azimuth and elevation). The cell is therefore expected to respond best to a rotation of the animal around the axis illustrated in the top left drawing, as will happen during flight. Adapted from Krapp and Hengstenberg (1996) Nature384: 463–466.

Figure 4.

(a) The knee‐jerk reflex relies on direct synaptic connections between sensory and motor neurons. Muscle spindles that sense the stretch of the quadriceps muscle resulting from tapping the knee tendon excite a pool of motor neurons in the spinal cord causing contraction of the quadriceps (agonist) and inhibit a pool of motor neurons that contract the antagonist muscle via interneurons. This information is conveyed to the brain by ascending pathways. More complex voluntary behaviours such as reaching are controlled by descending pathways conveying motor commands from the brain. (b) Coordinate transformations are essential for proper execution of motor programmes. If a subject fixes a cup directly (left panel), its position on the retina will be different than if they are looking at something else (right panel). The retinal position of the cup must therefore be combined with eye position information relative to the head to provide the appropriate motor command moving the arm. (c) The crayfish tail‐flip escape response illustrated at the bottom is mediated by a simple circuit comprising a pair of neurons on each side of the body called the lateral giant fibre (LG) and the medial giant fibre (MG). A single action potential in any of these neurons causes the activation of motor neurons in several body segments, resulting in the escape behaviour. Sensory neurons innervating LG and MG originate in part from different body segments. Interneurons play an important role in shaping the timing of the response. Adapted from Edwards et al. (1999) Trends in Neurosciences22: 153–161, with permission from Elsevier Science. (d) Response of a neuron in a premotor cortical area during a direction discrimination task. The monkey observes a motion signal consisting of dots moving in either of two directions and is trained to report the direction of motion by an eye movement after a delay. The particular neuron illustrated at the bottom shows strong activity during the delay period for an eye movement to the right but not for an eye movement to the left, thus revealing the intention of the monkey. The small bar on each graph indicates the occurrence of the eye movement. Adapted from Leon and Shadlen (1998) Neuron21: 669–672, with permission from Cell Press.

Figure 5.

(a) Anatomy and relative positioning of extraocular muscles. The muscular apparatus for eye movements consists of four recti (superior, inferior, lateral and medial) and two oblique muscles (inferior and superior). The superior oblique muscle passes through a pulley of bone called the trochlea. (b) Firing frequency of ocular motor neurons during fast eye movements and fixation. The top panel shows the action potentials recorded from a motor neuron during a fast horizontal eye movement, called a saccade, consisting of a 7.9° displacement of the eye following a 34 ms burst of spikes in the motor neuron. Note that the firing rate is constant before and after the movement and that it is higher after the movement than before it. The bottom panel illustrates the relationship between horizontal eye position and average firing rate in two different motor neurons. In both cases the relation between eye position and firing rate is linear, although the slopes and intercepts differ considerably. Adapted from Hepp et al. (1989) In: Wurtz and Goldberg (eds) The Neurobiology of Saccadic Eye Movements, pp. 105–212, Amsterdam: Elsevier. Copyright © 1989, with permission from Elsevier Science. (c) Forward swimming in the lamprey is accomplished through an undulatory rhythmic movement of the body from head to tail, which requires the coordination of muscle activity in successive body segments. As illustrated on the left, the phase of muscle activation is delayed from segment to segment during the swim cycle. Furthermore, in each segment, muscles contracting the body on the left and right side are activated in alternation. (d) Schematic diagram of the neuronal network (central pattern generator) located in the spinal cord which generates the motor pattern activating body muscles in each segment. On each body side a pool of excitatory interneurons (E) activates motor neurons (M) causing body contraction and simultaneously inhibits neurons on the opposite side via a pool of inhibitory interneurons (I). This mechanism ensures the alternate contractions of muscles on both sides during swimming. Sensory feedback from stretch receptors on each side of the body and the activation of lateral interneurons (L) also contribute to the motor pattern. Adapted from Grillner et al. (1995) Trends in Neurosciences18: 270–279. Copyright © 1995, with permission from Elsevier Science.

close

References

Alle H and Geiger JRP (2008) Analog signalling in mammalian cortical neurons. Current Opinion in Neurobiology 18(3): 314–320.

Angelaki D, Gu Y and DeAngelis GC (2011) Visual and vestibular cue integration for heading perception in extrastriate visual cortex. Journal of Physiology 589: 825–833.

Barlow HB, Levick WR and Yoon M (1971) Responses to single quanta of light in retinal ganglion cells of the cat. Vision Research 3(suppl.): 87–101.

Block SM (1992) Biophysical principles of sensory transduction. Chap. 1. In: Corey DP and Roper SD (eds) Sensory Transduction, Vol. 47, Society of General Physiologists Series, pp. 1–17. New York: Rockefeller University Press.

Bogacz R, Wagenmakers EJ, Forstmann BU and Nieuwenhuis S (2009) The neural basis of the speed‐accuracy tradeoff. Trends in Neurosciences 33: 10–16.

Borst A, Haag J and Reiff D (2010) Fly motion vision. Annual Review of Neuroscience 33: 49–70.

Bullock TH and Heiligenberg W (1986) Electroreception. New York: John Wiley & Sons.

Burgess N and O'Keefe J (2011) Models of place and grid cell firing and theta rhythmicity. Current Opinion in Neurobiology 21: 734–744.

Buzsáki G and Draguhn A (2004) Neuronal oscillations in cortical networks. Science 304: 1926–1929.

Churchland PS and Sejnowski TJ (1992) The Computational Brain. Cambridge, MA: MIT Press.

Crapse TB and Sommer MA (2008) Corollary discharge across the animal kingdom. Nature Reviews Neuroscience 9: 587–600.

Derdikman D and Moser EI (2010) A manifold of spatial maps in the brain. Trends in Cognitive Sciences 14: 561–569.

Deschênes M, Moore J and Kleinfeld D (2011) Sniffing and whisking in rodents. Current Opinion in Neurobiology 22: 1–8.

Domenici P, Booth D, Blagburn JM and Bacon JP (2008) Cockroaches keep predators guessing by using preferred escape trajectories. Current Biology 18: 1792–1796.

Ermentrout GB, Galán RF and Urban NN (2008) Reliability, synchrony and noise. Trends in Neurosciences 31: 428–434.

Faisal AA, Selen LPJ and Wolpert DM (2008) Noise in the nervous system. Nature Reviews Neuroscience 9: 292–303.

Fotowat H and Gabbiani F (2011) Collision detection as a model for sensory‐motor integration. Annual Review of Neuroscience 34: 1–19.

Friedrich RW (2006) Mechanisms of odor discrimination: neurophysiological and behavioral approaches. Trends in Neurosciences 29: 40–47.

Gács P (1986) Reliable computation with cellular automata. Journal of Computer and System Sciences 32: 15–78.

Gollisch T and Meister M (2010) Eye smarter than scientists believed: neural computations in circuits of the retina. Neuron 65: 150–164.

Gould JL (2010) Magnetoreception. Current Biology 20(10): R431–R435.

Gu Y, Oberwinkler J, Postma M and Hardie RC (2005) Mechanisms of light adaptation in Drosophila photoreceptors. Current Biology 15(13): 1228–1234.

Homberg U, Heinze S, Pfeiffer K et al. (2011) Central neural coding of sky polarization in insects. Philosophical Transactions of the Royal Society B 366: 680–687.

Juusola M, French AS, Uusitalo RO and Weckström M (1996) Information processing by graded‐potential transmission through tonically active synapses. Trends in Neurosciences 19: 292–297.

Kerhuel L, Viollet S and Franceschini N (2012) The VODKA sensor: a bio‐inspired hyperacuity optical position sensing device. IEEE Sensors Journal 12: 315–324.

King AJ (2005) Multisensory integration: strategies for synchronization. Current Biology 15: R339–R341.

Krahe R and Gabbiani F (2004) Burst firing in sensory systems. Nature Reviews Neuroscience 5: 13–23.

Llinás RR (1988) The intrinsic electrophysiological properties of mammalian neurons: insights into central nervous system function. Science 242: 1654–1664.

Ma WJ, Beck JM and Pouget A (2008) Spiking networks for Bayesian inference and choice. Current Opinion in Neurobiology 18: 217–222.

Masland RH and Martin PR (2007) The unsolved mystery of vision. Current Biology 17: R577–R582.

McLaughlin T and O'Leary DDM (2005) Molecular gradients and development of retinotopic maps. Annual Review of Neuroscience 28: 327–355.

Moss CF and Surlykke A (2010) Probing the natural scene by echolocation in bats. Frontiers in Behavioral Neuroscience 4(33): 1–16.

Nienborg H and Cumming B (2010) Correlations between the activity of sensory neurons and behavior: how much do they tell us about a neuron's causality? Current Opinion in Neurobiology 20: 376–381.

Niven JE and Laughlin SB (2008) Energy limitation as a selective pressure on the evolution of sensory systems. Journal of Experimental Biology 211: 1792–1804.

O'Carroll DC, Bidwell NJ, Laughlin SB and Warrant EJ (1996) Insect motion detectors matched to visual ecology. Nature 382(6586): 63–66.

Oláh S, Füle M, Komlósi G et al. (2009) Regulation of cortical microcircuits by unitary GABA‐mediated volume transmission. Nature 461: 1278–1281.

Padmanabhan K and Urban NU (2010) Intrinsic biophysical diversity decorrelates neuronal firing while increasing information content. Nature Neuroscience 13: 1276–1282.

Palmer CR and Kristan WB (2011) Contextual modulation of behavioral choice. Current Opinion in Neurobiology 21: 520–526.

Philippides A, Ott SR, Husbands P, Lovick TA and O'Shea M (2005) Modeling cooperative volume signaling in a plexus of nitric oxide‐synthase expressing neurons. Journal of Neuroscience 25: 6520–6532.

Rivera‐Alba M, Vitaladevuni SN, Mishchenko Y et al. (2011) Wiring economy and volume exclusion determine neuronal placement in the Drosophila brain. Current Biology 21: 2000–2005.

de Ruyter van Steveninck RR and Laughlin SB (1996) The rate of information transfer at graded‐potential synapses. Nature 379: 642–645.

Schmidt TM, Chen SK and Hattar S (2011) Intrinsically photosensitive retinal ganglion cells: many subtypes, diverse functions. Trends in Neurosciences 34(1): 572–580.

Sharpeshkar R (1998) Analog versus digital: extrapolating from electronics to neurobiology. Neural Computation 10: 1601–1638.

Smear M, Shusterman R, O'Connor R, Bozza T and Rinberg D (2011) Perception of sniff phase in mouse olfaction. Nature 479: 397–400.

Speed T (2011) A Correlation for the 21st century. Science 334: 1502–1503.

Thomson EE and Kristan WB (2005) Quantifying stimulus discriminability: a comparison of information theory and ideal observer analysis. Neural Computation 17: 741–778.

Varga Z, Jia H, Sakmann B and Konnerth A (2011) Dendritic coding of multiple sensory inputs in single cortical neurons in vivo. Proceedings of the National Academy of Sciences of the USA 108: 15420–15425.

Wachowiak M (2011) All in a sniff: olfaction as a model for active sensing. Neuron 71: 962–973.

Westheimer G (2009) Hyperacuity. In: Squire LR (ed) Encyclopedia of Neuroscience. Oxford: Academic Press.

White JG, Southgate E, Thomson JN and Brenner S (1986) The structure of the nervous system of the nematode Caenorhabditis elegans. Philosophical Transactions of the Royal Society B: Biological Sciences 314: 1–340.

Wolfe J, Houweling AR and Brecht M (2010) Sparse and powerful cortical spikes. Current Opinion in Neurobiology 20: 306–312.

Further Reading

Bradbury JW and Vehrencamp SL (2011) Principles of Animal Communication, 2nd edn. Sunderland, MA: Sinauer Associates.

Chittka L and Niven J (2009) Are Bigger Brains Better? Current Biology 19: R995–R1008.

Cover TM and Thomas JA (2006) Elements of Information Theory, 2nd edn. Hoboken, NJ: John Wiley & Sons.

Franklin DW and Wolpert DM (2011) Computational mechanisms of sensorimotor control. Neuron 72: 425–442.

Gabbiani F and Cox SJ (2010) Mathematics for Neuroscientists. San Diego: Academic Press.

Johansson S and Flanagan JR (2009) Coding and use of tactile signals from the fingertips in object manipulation tasks. Nature Reviews Neuroscience 10: 345–359.

Marder E (2011) Variability, compensation, and modulation in neurons and circuits. Proceedings of the National Academy of Sciences of the USA 108: 15542–15548.

Selverston A (2010) Invertebrate central pattern generator circuits. Philosophical Transactions of the Royal Society B 365: 2329–2345.

Silver RA (2010) Neuronal arithmetic. Nature Reviews Neuroscience 11: 474–489.

Zupanc GKH (2010) Behavioral Neurobiology, 2nd edn. New York: Oxford University Press.

Contact Editor close
Submit a note to the editor about this article by filling in the form below.

* Required Field

How to Cite close
Gabbiani, Fabrizio, and Midtgaard, Jens(Oct 2012) Neural Information Processing. In: eLS. John Wiley & Sons Ltd, Chichester. http://www.els.net [doi: 10.1002/9780470015902.a0000149.pub2]