* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Lecture
Endocannabinoid system wikipedia , lookup
Recurrent neural network wikipedia , lookup
Environmental enrichment wikipedia , lookup
Cognitive neuroscience of music wikipedia , lookup
Binding problem wikipedia , lookup
Mirror neuron wikipedia , lookup
Neuropsychology wikipedia , lookup
Brain Rules wikipedia , lookup
Multielectrode array wikipedia , lookup
Neurogenomics wikipedia , lookup
Time perception wikipedia , lookup
Neuroinformatics wikipedia , lookup
Functional magnetic resonance imaging wikipedia , lookup
Artificial general intelligence wikipedia , lookup
Biochemistry of Alzheimer's disease wikipedia , lookup
Human brain wikipedia , lookup
Electrophysiology wikipedia , lookup
Types of artificial neural networks wikipedia , lookup
Neuroesthetics wikipedia , lookup
Central pattern generator wikipedia , lookup
Neurotransmitter wikipedia , lookup
Haemodynamic response wikipedia , lookup
Nonsynaptic plasticity wikipedia , lookup
Neural engineering wikipedia , lookup
Aging brain wikipedia , lookup
Premovement neuronal activity wikipedia , lookup
Neural oscillation wikipedia , lookup
Single-unit recording wikipedia , lookup
Cognitive neuroscience wikipedia , lookup
Neuroplasticity wikipedia , lookup
Neuroeconomics wikipedia , lookup
Synaptogenesis wikipedia , lookup
Neural coding wikipedia , lookup
Circumventricular organs wikipedia , lookup
Activity-dependent plasticity wikipedia , lookup
History of neuroimaging wikipedia , lookup
Pre-Bötzinger complex wikipedia , lookup
Neurophilosophy wikipedia , lookup
Biological neuron model wikipedia , lookup
Chemical synapse wikipedia , lookup
Holonomic brain theory wikipedia , lookup
Stimulus (physiology) wikipedia , lookup
Development of the nervous system wikipedia , lookup
Optogenetics wikipedia , lookup
Neural correlates of consciousness wikipedia , lookup
Clinical neurochemistry wikipedia , lookup
Spike-and-wave wikipedia , lookup
Molecular neuroscience wikipedia , lookup
Neuroanatomy wikipedia , lookup
Synaptic gating wikipedia , lookup
Feature detection (nervous system) wikipedia , lookup
Channelrhodopsin wikipedia , lookup
Metastability in the brain wikipedia , lookup
Neural Information Processing: Normal and Pathological Péter Érdi Center for Complex Systems Studies, Kalamazoo College, Kalamazoo, MI 49006 and Department of Biophysics, KFKI Research Institute for Particle and Nuclear Physics of the Hungarian Academy of Sciences Budapest, Hungary last revised: September 23, 2003 [email protected] Comments for CCSS specific pages 2 KFKI RMKI Computational Neuroscience Group http://www.rmki.kfki.hu/biofiz/cneuro/ The Computational Neuroscience Group Department of Biophysics KFKI Research Institute for Particle and Nuclear Physics of the Hungarian Academy of Sciences Budapest, Hungary H-1121 Budapest, Konkoly-Thege út 29-33 Mail address: P.O.Box 49, H-1525 Budapest, Hungary Phone: +36 (1) 392-2222 Fax: +36 (1) 392-2742 Welcome to our homepage. We hope you will enjoy our pages and wish you good luck for navigating through them. Although we try to do our best to make this site as interesting, fascinating and up-to-date as possible and at the same time also easy to handle and hard to get lost in, still you must keep in mind that the contents are under construction, constant rewriting and reediting so you might meet some errors or signs of uncompleteness - for which we apologize. About the group People Projects Recent publications Lecture materials Seminars, symposia, conferences Education International cooperations Other neuroscience sites Selected Topics (Download) 1 of 2 09/21/04 15:24 3 Contents • BRAIN: Methods, Levels, Representations, Computation • NEUROLOGICAL DISORDERS AS DYNAMICAL DISEASES • ANXIETY • TOWARDS A COMPUTATIONAL/PHYSIOLOGICAL MOLECULAR SCREENING and DRUG DISCOVERY 4 BRAINS: Methods, levels, representations, computation 5 Brains • Experimental methods and disciplines • Levels • Neural representation: cells, networks, modules • Neural computation versus computational neuroscience 6 Experimental methods and disciplines • Anatomy: discover the structure of the brain • Electrophysiology – Single cell recording – Electroencephalography (EEG) • Psychological experiments on people with brain damage, e.g. HM • Brain imaging 7 Brain imaging • EEG and MEG • CT or CAT: Computerized axial tomography • PET : Positron emission tomography • fMRI: Functional magnetic resonance imaging 8 EEG and MEG EEG is an acronym for Electroencephalograph This is a recording (”graph”) of electrical signals (”electro”) from the brain (”encephalo”) electroencephalographs from humans: Hans Berger (1929) The skull has low conductivity to electric current but is transparent to magnetic fields, the measurement sensitivity of the magnetoencephalography (MEG) in the brain region should be more concentrated than that of the electroencephalography (EEG). BRAIN WAVES normal, pathological EEG and MEG: relatively good temporal resolution (∼ms), poor spatial resolution detection of lesions 9 Computerized axial tomography • A computerized axial tomography scan: CAT scan or CT scan • It is an x-ray procedure which combines many x-ray images with the aid of a computer to generate cross-sectional views and • 3D images of the internal organs and structures • CAT scan is used to define normal and abnormal structures in the body • Nobel prize in medicine: 1979 • Hounsfield and Cormack 10 Positron emission tomography “hemodynamic” changes first scanning method to give FUNCTIONAL information on the brain measures the emission of positrons from the brain after a small amount of radioactive isotopes, or tracers, have been injected into the blood stream. A common example is a glucose-relative with embedded fluor-18. With this molecule, the activity of different regions of the brain can be measured. The result is a three-dimentional map with the brain activity represented by colors. function 11 Functional magnetic resonance imaging • Head is inserted into a giant circular magnet • The magnet realigns atoms and protons • Machine sends a radio pulse that causes atoms to release energy • A computer detects energy release and produces an image • Functional MRI measures brain function by measuring changes in blood volume • Possible to map functions down to a few millimeters Sample results: http://www.fmrib.ox.ac.uk/ 12 Functional magnetic resonance imaging Limitations of fMRI • Makes strong assumptions about relation between brain activity and blood flow • Numerous calculations required • Low temporal resolution because of sluggish metabolic response 13 LEVELS 14 Levels 15 Hierarchy of dynamics • Neuron dendrites axon soma (cell body) + or − • Synapse I(t) O(t) • Network interneurons afferent neurons efferent neurons 16 Hierarchy of dynamics Neuron and synapse Neuron complex morphology membrane potential dynamics dendrites signal generating and processing device axon soma (cell body) stimulus signal Synapse ° structural and functional connection between cells specialized for trasferring information ° ° ° ° neurochemically mediated synapses (mostly) excitatory and inhibitory synaptic strength (synaptic weight, synaptic efficacy) d s ij (t) dt + or − learning 17 Hierarchy of dynamics Network Network I(t) O(t) interneurons afferent neurons d s ij (t) = 0 dt efferent neurons fixed wiring pattern formation problems d s ij (t) dt 0 pattern recognition problems 18 Large-scale organization 19 Large-scale organization Cortical Area Function Prefrontal Cortex Problem Solving, Emotion, Complex Thought Motor Association Cortex Coordination of complex movement Primary Motor Cortex Initiation of voluntary movement Primary Somatosensory Cortex Receives tactile information from the body Sensory Association Area Processing of multisensory information Visual Association Area Complex processing of visual information Visual Cortex Detection of simple visual stimuli Wernicke’s Area Language comprehension Auditory Association Area Complex processing of auditory information Auditory Cortex Detection of sound quality (loudness, tone) Speech Center (Broca’s Area) Speech production and articulation 20 Neural representation: cells, networks, modules Representation with neurons and populations of neurons A. Resting membrane potential. B. If voltage change surpasses this value, the depolarization automatically goes to its maximum value. C. If a stimulus arrives during this phase of the potential it can not stimulate the neuron. D. Only a very strong stimulus can stimulate the neuron during this phase. E. Resting potential. F. Absolute refractory period. G. Relative refractory period. 21 Neural representation: cells, networks, modules Representation with neurons and populations of neurons A typical neuron can fire as much as 100 times per second. Spike train of a neuron: its pattern of firing or not firing over a period of time. 10100 and 00011: both involve a neuron with a firing rate of 2 times out of 5 ”rate code” vs. ”temporal code” Is there, at some high level, a cell so specialized that it only signals when viewing a particular object? This is the idea of grandmother cells: a single cell might be responsible for signalling the presence of a single stimulus, for example, your grandmother. sharp debate on ”grandmother cell” 22 Neural representation: cells, networks, modules Representation with neurons and populations of neurons II. Do we really have a certain nerve cell for recognising the concatenation of features representing our grandmother(s)? Population (ensemble) code: Perception depends on the combined output of a group (ensemble) of cells not on the ouput of any one cell in particular. Both natural and most artificial neural networks use distributed representations oncepts are encoded by a population of neurons: a group of neurons represents a concept by virtue of a pattern of firing rates in all of the neurons. Thus a group of neurons, each with its own firing rate, can encode a large number of aspects of the world. 23 Neural representation: cells, networks, modules Modules of the visual cortex Different brain regions represent different kinds of sensory stimuli visual cortex: has neurons that respond to different visual inputs orientation selectivity, position selectivity, colour selectivity, direction selectivity the response of each of these cells is dominated by one eye binocular integration: ocular dominance columns 24 Neural computation Is the brain a computer? computation: transformation of representations 2+3=5 p->q, p; so q Brain: performs transformations of representations encoded by the firing patterns of neurons Retinal cells: respond to light reflected from objects into your eye, and send signals through a series of layers of neurons in the visual cortex. Successive layers: detect more and more complex aspects of the objects that originally sent light into the eye, as neurons in each layer abstract and transform the firing patterns of neurons in the preceding layer Visual cortex: progressively constructs representations of lines, patterns in two dimensions, and finally three-dimensional colored objects 25 GENETIC versus DYNAMICAL DISEASES genetic disease: caused by abnormalities in an individuals genetic material (genome). There are four different types of genetic disorders: • single-gene • multifactorial • chromosomal • mitochondrial even genetic disorders have dynamic aspects 26 NEUROLOGICAL DISORDERS AS DYNAMICAL DISEASES I Dynamical disease occurs due to the impairment of the control system: associated to ‘abnormal’ dynamics • Develop realistic mathematical models and study effects of parameter changes • Neurobiological interpretation • Integration of molecular, cellular and system neuroscience • Therapeutic strategies 27 NEUROLOGICAL DISORDERS AS DYNAMICAL DISEASES II CHANGES in Qualitative Dynamic States state state time time Parkinson disease memory performance synaptic loss Epilepsy Alzheimer disease 28 EPILEPSY Cellular and Network Mechanisms of Epileptogenesis 1. epileptic neuron hypothesis: • altered intrinsic firing properties • individual neurons • spontaneous bursting 2. epileptic aggregate hypothesis: • aberrant connectivity • recurrent excitations • normal neurons are driven into burst Probably both are true 29 • Typical single cell effects: many possible changes in dynamics of ion-currents can result similar pathological activity. membrane potential[mV] EPILEPSY soma • Increasing of Ca2+ conductance can transform a simple spiker cell to burster. membrane potential[mV] time[msec] soma time[msec] 30 EPILEPSY • Epileptic seizures in hippocampus and in certain cortical areas E1 • Large amplitide, quick waves in EEG records E2 I1 Excitation GABAergic inhibition [bicucullin, penicillin] I2 1. oversynchronization 2. irregularity Antiepyleptic drugs to influence GABAA transmission. 31 EPILEPSY • predicting epileptic seizures (earthquakes, stock market crashes?) • seizure control (feedback, electric fields: S. Schiff) • general theory? (Milton and Jung) 32 PARKINSON DISEASE (hypokinesia, dyskinesia) • Decrease of DOPA neuron’s number (approx. 80%) in Substantia Nigra. • Examination of the effect of DOPA using neuron network models. excitation DOPA → decrease threshold fixed point attractor periodic attractor • Modification of Central Pattern Generators 33 SCHIZOPHRENIA Positive and negative symptoms state uncomplicated actions and speech decreased motivation state hallucination Time Time ’waving’ ’steady’ ’E’ storage and recall of memory traces ’E’ Models: • ‘lesion models’ does not explain waving • neurotransmitter model (DOPA) • disconnection hypothesis • NMDA: delayed maturation of NMDA receptors • cortical pruning (synaptic depression) state state changes in attractor structure pathological attractors 34 SCHIZOPHRENIA THE NMDA RECEPTOR DELAYED MATURATION HYPOTHESIS Pathological attractors appear recall of never learned items ’E’ ’E’ recall of learned memory traces state state ’delusion’ ’hallucination’ 35 Anxiety A disorder characterized by generalized and persistent free-floating anxiety (anxiety not restricted to any particular event or circumstance). The symptoms are variable, and can include muscle tension, continuous feelings of nervousness, trembling, sweating, lightheadedness, dizziness, palpitations. A variety of worries are often expressed, including the worry that the sufferer or a relative will have an accident or become ill. Sleep disturbance is common. Generalized Anxiety Disorder is more common in women than men. One factor in its development appears to be chronic stress. Its course varies, but tends to be fluctuating and chronic. 36 Overview SEPTOHIPPOCAMPAL SYSTEM THETA RHYTHM SKELETON NETWORK i (b) pyr i (O−LM) i (S) KNOCK−IN, KNOCK−OUT TECHNIQUES MS−GABA DESENSITIZATION KINETICS RECEPTOR SUBUNITS GABA SYNAPSE 37 The septohippocampal system • the medial septal region and the hippocampus are connected reciprocally via GABAergic neurons, • Toth K, Borhegyi Z, Freund TF. J Neurosci. 1993 • the physiological role of this loop is still not well understood 38 The GABAA receptor Subunit composition Subunit composition of synaptic GABAA receptors GABAA receptors are pentameric heterooligomers assembled from members of seven different subunit classes, some of which have multiple members: α(1-6), β(1-3), γ(1-3), δ, , θ and π. 39 The GABAA receptor Subunit composition Complex patterns of subunit distribution in different brain regions and cell types support the view that dozens of GABAA receptor combinations are present in the brain. 40 TOWARDS A COMPUTATIONAL/PHYSIOLOGICAL MOLECULAR SCREENING (and DRUG DISCOVERY) Desired temporal pattern Nontrivial enhanced cognition e.g. theta: anxiogenics Septohippocampal system Temporal pattern computational & pharmaceutical modulation Comp. interface to further testing 41 Physiological Experiments Theta activity can be depressed and induced in the hippocampal CA1 by GABA allosteric modulators The positive allosteric modulator diazepam, which increases chloride conductance, suppresses theta activity Negative allosteric modulator FG7142, which reduces transmission at GABA-A synapses, reverses the effect of diazepam 42 The skeleton network i (b) pyr i (O−LM) i (S) MS−GABA 43 Theta modulation in the MS-CA1 system negative allosteric modulator control A B 10 pyr n=10 5 0 1 1.5 2 2.5 3 3.5 5 0 4 45 15 5 1.5 2 2.5 3 3.5 i (o/a) n=30 20 15 5 1 1.5 2 2.5 3 25 3.5 4 MS−GABA n=50 20 2 2.5 3 3.5 15 5 1 1.5 2 2.5 3 3.5 4 25 Effect of negative allosteric modulator was taken into account by lower synaptic conductance at all pathways i (o/a) n=30 20 10 5 0 1 1.5 2 2.5 3 25 15 10 3.5 4 MS−GABA n=50 20 10 5 4 i (b) n=100 25 15 0 1.5 15 10 0 1 35 4 25 o f o f 1 n u m b e r c e l l s i (b) n=100 25 n u m b e r c e l l s 45 35 pyr n=10 10 In all neuron populations clustering of spikes occurs at lower synaptic conductance values 5 1 1.5 2 2.5 3 3.5 4 0 1 1.5 time [s] 2 2.5 3 3.5 4 time [s] t EEG power [mV²] C Timing of action potentials tends to have a well defined value 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 5 10 15 20 25 frequency [Hz] 30 t Theta power in EEG computed from the activity of pyramidal neurons shows a significant increase during simulated administration of the negative allosteric modulator 44 CONCLUSIONS DYNAMICAL MODELS HIERARCHICAL LEVELS INTEGRATION of LEVELS THERAPEUTIC STRATEGIES 45