Neural Basis of Memory A CS781 termpaper Presentation
Awhan Patnaik Swagat Kumar
Department of Computer Science and Engineering, Indian Institute of Technology, Kanpur. November 7, 2004
Memory and Cognition
Page 1
Motivation
• Are neurons the loci of memory? • Are memories localized or distributed? • How does the transmission of signal take place in neurons? • Do genes play a role in memory formation? • How information is encoded in neural pulses? • Do we have an answer to the Binding problem?
November 7, 2004
Memory and Cognition
Page 2
Contents
• Introduction • The Neuron • Signal Transmission in Neurons • Genetic study of Memory • Information encoding in Neural Pulses • Unified Cognition • Summary November 7, 2004
Memory and Cognition
Page 3
Introduction
• Memory and Neuron: The Connection. • Encoding • Storage • Retrieval
November 7, 2004
Memory and Cognition
Page 4
Introduction
Memory & Neurons: The Connection • Evidence – STM disruption - blow on the head, epileptic seizure etc → trouble remembering – Brain Lesions - Lashley’s Mice, the patient H.M. – PET scans,fMRI data – Knock Out Mice - defective fyn gene (Philippe Soriano) As a scientist I am compelled to the conclusion-not postulation, not assumption, but conclusion that there must exist certain physical-chemical changes in the nervous tissue that correspond to the storage of information, or to the engram, changes that constitute the necessary conditions of remembering (Tulving, cited in Gazzaniga,1997, p. 97).
November 7, 2004
Memory and Cognition
Page 5
Introduction
Encoding, Storage & Retrieval
Encoding: internal representations of external stimuli. Storage: distributed yet localized Retrieval: associative pattern matching - cued recall
November 7, 2004
Memory and Cognition
Page 6
The Neuron
• Structure • Resting Potential • Action Potential • Signal transmission
November 7, 2004
Memory and Cognition
Page 7
The Neuron
Structure
Figure 1: A Neuron
November 7, 2004
Memory and Cognition
Page 8
The Neuron
Resting Potential • at rest the inside of the neuron is -ve wrt. to its outside • cell membrane is selectively permeable • at rest Ka+ ions can pass thru but Cl− & N a+ have difficulty whereas -vely charged protein molecules cannot pass through selective ion channels • relatively more sodium ions outside the neuron and more potassium ions inside that neuron • a pump to move three sodium ions out of the neuron for every two potassium ions it puts in • typical value of resting potential ∼ -70mV
November 7, 2004
Memory and Cognition
Page 9
The Neuron
Action Potential
Figure 2: Action Potential
November 7, 2004
Memory and Cognition
Page 10
The Neuron
• stimulus causes N a+ ion channels to open • more N a+ ions outside, so N a+ ions rush in depolarizing the neuron • when the depolarization reaches about -55 mV a neuron will fire an action potential • it takes longer for K + channels to open, when they do open, K + ions rush out reversing the depolarization • at about this time N a+ channels start to close causing a repolarization towards -70mV • there is actually a hyperpolarization going past -70mV because the K + channels stay open a bit too long • gradually ion channels go back to resting levels
November 7, 2004
Memory and Cognition
Page 11
The Neuron
Signalling: Chemical Synapse • at rest the neuron is negative w.r.t outside • Neurotransmitters stored at axon tips in synaptic vesicles • incoming AP at the axon tip causes Ca2+ channels to open up • influx of Ca2+ ions triggers exocytosis of some of the vesicles which releases neurotransmitters in the synaptic cleft • Neurotransmitters may be excitory or inhibitory – excitory - acetylcholine – inhibitory - GABA, glycine • if temporal summation of incoming nerve impules, exceeds a threshold (∼-55mV) the synaptic vesicles diffuse and relase the neurotransmitters by exocytosis into the synaptic cleft • once in the synapse, neurotransmitters are active for only a short time ∼ 0.5-1 November 7, 2004
Memory and Cognition
Page 12
The Neuron milliseconds • on reaching the post-synaptic dendrites the neurotransmitters bind themselves to receptors • receptors are ligand-gated ion channels • two general categories of receptor proteins: – ionotropic: Activation of ionotropic receptors causes membrane ion channels to open or close. – metabotropic: Activation of metabotropic receptors involves an intracellular biochemical cascade which may end with the opening or closing of ion channels or other intracellular effects. • Excitory neurotransmitters make it easier for the cell to allow positive ions in, and therefore decrease the threshold • Inhibitory neurotransmitter, on the other hand, make the neuron’s membrane more permeable to negative ions, and increase the threshold. November 7, 2004
Memory and Cognition
Page 13
Genetic Study of Memory
• Introduction • Hebb’s Hypothesis • Memory in Invertebrates • Long Term Potentiation
November 7, 2004
Memory and Cognition
Page 14
Genetic Study of Memory
Introduction
• Is memory a distinct property of mind? Can memory be studied independent of other higher cognitive functions? Answer: memory is indeed a distinctive cognitive function that can be studied independent of other higher cognitive abilities. Evidence: H.M. Case Study. Surgical removal of a part of his medial temporal lobe led to severe memory deficit but it did not affect cognitive functions.
November 7, 2004
Memory and Cognition
Page 15
Genetic Study of Memory
• If memory can be isolated and studied as an independent process, is there any reason to think that there will be genes that are specifically devoted to memory? Answer: It is less clear. – Most of genes so far identified as affecting memory are involved in signal transduction pathways that are recruited for purposes unrelated to memory in other cell types. – However, given that memory is stored in certain specific structures in the brain, it is likely that some isoforms exist that affect these structures selectively either developmentally or for mature cellular function. – conditional genetic approaches allow one to target gene modifications selectively to structures specifically involved in memory, such as hippocampus.
November 7, 2004
Memory and Cognition
Page 16
Genetic Study of Memory
Hebb’s Hypothesis Donald Hebb proposed that when a receptor in the forebrain or hippocampus is hit by a glutamine neurotransmitter, not only is the signal received and passed along, but an acknowledgment is sent back to the nerve cell that threw the neurotransmitter. This receipt strengthens the connection between the two nerve cells, making future communication easier. Any nerve cell junction in the neighborhood that happened to be firing at just that time gets strengthened even ones that had nothing to do with the original message. Thus nerve cells that fire at the same time tend to establish firm junctions with one another. This process of linking together nerves that fire at the same time is called association. He hypothesized that the learning is the result of molecular interactions that take place between the nerve cells in the brain. November 7, 2004
Memory and Cognition
Page 17
Genetic Study of Memory
Memory in Invertebrates Evidences for involvement of genes in memory formation 1. 1970s, Seymour Benzer’s group at Cal Tech: Used genetic screening to identify four kinds of drosophila mutants (dunce, rutabaga, amnesiac and linotte) which could not learn odor-shock conditioning task. Three of them had a defective molecule involved in cAMP signalling. 2. 1970, Kandel’s group: In aplysia, under tail shock-gill withdrawal conditioning, the sensory siphon withdrawal. The sensory-motor synapse can undergo 3 different stages of facilitation depending on the number of behavioral training trials. • single pulse of 5-HT activates pathway 1 leading to facilitation that lasts few minutes. Pathway 1: stimulus → Serotonin → Adenylate cyclase → cAMP → PKA → Phosphorylation of K+ chanel → gill withdrawal. November 7, 2004
Memory and Cognition
Page 18
Genetic Study of Memory
• Multiple pulses of 5-HT activates pathway 2 and 3 where cAMP dependent protein kinase translocates to nucleus and activates transcription via CREB pathway. Pathway 2: stimulus → Serotonin → Adenylate cyclase → cAMP → PKA → CREB → Ubiquitin Hydrolase → persistently active kinase (proteolysis) → 12-24 hours facilitation. • pathway 3: stimulus → Serotonin → Adenylate cyclase → cAMP → PKA → CREB → C/EBP transcription factor → new synapse → prolonged stabilization of synaptic facilitation. 3. Pramod Das, Yin and Tully, 1990: Experiments on drosophilla and aplysia revealed following facts: • long-term memory is blocked by the administration of protein synthesis inhibitors during learning while short-term memory is resistant to protein synthesis inhibition. This indicates that new genes are expressed during learning which are necessary for long-term memories. November 7, 2004
Memory and Cognition
Page 19
Genetic Study of Memory • cAMP responsive transcription factor CREB is critically involved in conversion of short-term to long-term memory. • Long term memory may be encoded in the pattern of strengthening of synaptic connections between neurons. • CREB activation is the rate limiting switch for conversion of short-term to long-term memory. Expression of the dCREB2 transgene can overcome this requirement.
What do the experiments on invertebrates tell us? •
cAMP signaling pathway can be used to change the strength of synaptic connections between neurons and that this alteration in connectivity is necessary to form memories.
• A CREB mediated induction of transcription is necessary to produce the long-lasting changes in synaptic strength required for the long-term storage of memories. November 7, 2004
Memory and Cognition
Page 20
Genetic Study of Memory
Long Term Potentiation It is a process in which synapses are strengthened. Various steps involved in LTP: • Intense stimulation of pre-synaptic neuron leads to sufficient depolarization of post-synaptic neuron which evacuates magnesium ions (M g 2+ ) blocking the NMDA receptors. • This allows calcium ions (Ca2+ ) to enter the dendrites. • Ca2+ activates one of the enzymes called calmoduline by combining with it and forming Ca2+ /calmodulin. • Ca2+ /calmodulin in turn activates other enzymes like adenylate cyclase and CaM Kinase II which lead to phosphorylation of other molecules. • The activated adenylate cyclase manufactures cyclic adenosine November 7, 2004
Memory and Cognition
Page 21
Genetic Study of Memory
monophosphate (cAMP) which in turn catalyzes the activity of another protein PKA. it leads to cascade of biochemical reactions: – PKA phosphorylates AMPA receptors allowing them to remain open longer when glutamate binds to them. Thus post-synaptic neuron becomes further depolarized, contributing to LTP. – PKA activates CREB which plays a major role in gene transcription that leads to the creation of new AMPA receptors that increase the synaptic efficiency even further. • caMKII has a property of phosphorylating itself. It is essential for the persistence of LTP. Its activity continues long after the Ca2+ are evacuated and Ca2+ /calmodulin has been deactivated. • CaMKII in turn phosphorylates AMPA receptors and MAPK which are involved in the formation of dendrites or increasing the Ca2+ conductance of NMDA receptors. November 7, 2004
Memory and Cognition
Page 22
Genetic Study of Memory
Figure 3: Bio-Chemical reactions at a synapse during LTP
November 7, 2004
Memory and Cognition
Page 23
Genetic Study of Memory
LTP involves two phases: • Establishment phase: – Last for few hours. – Induced by single high frequency stimulus. – Involves activity of various enzymes which may persist even after the calcium is eliminated. – No protein synthesis. • Maintenance phase: – May persist for several days. – Requires a series of high-frequency stimuli. – Requires synthesis of new proteins. – Formation of new receptors, synapses. November 7, 2004
Memory and Cognition
Page 24
Information Encoding in Neural Systems • Receptive Fields • Single Neuron Codes • Population Neuronal Codes • Temporal Binding • Synchronization and Oscilation.
November 7, 2004
Memory and Cognition
Page 25
Information Encoding in Neural Systems
Receptive Fields • A neuron respond to a specific set of input patterns, known as its receptive field. • Highest level receptive fields are quite specific: grandmother cell hypothesis – We do not have enough neurons to code for every feature of every visual scene we encounter in our lives, so neurons must code for some level of general features. – individual neurons may die during our lifetimes, so there must be multiple neurons coding for similar features for robustness.
November 7, 2004
Memory and Cognition
Page 26
Information Encoding in Neural Systems
Single neuron Codes • Output of a neuron is a sequence of action potentials (APs). • The temporal pattern of action potential firing can be very different between neuronal types and for different states of the same neuron. • How might information be encoded by APs? – Binary – Rate encoding: average firing frequency over some time period. – Interspike Interval: temporal sequence of exact spike times. – Dynamic range: The number of states of a neuron at a given time is very large.
November 7, 2004
Memory and Cognition
Page 27
Information Encoding in Neural Systems
Figure 4: Action Potential temporal firing patterns
November 7, 2004
Memory and Cognition
Page 28
Information Encoding in Neural Systems
Population Neuronal Codes The Output of a network is a pattern of activity across the population of neurons and the information may be encoded in three different ways: • local coding: each neuron represents a specific feature that the system distinguishes • scalar coding: firing rate of each neuron encodes a feature – redundancy and improved signal-to-noise ratio can be achieved by several neurons coding the same features • vector coding: feature encoded in the firing rates of a subpopulation of neurons that have overlapping tuning curves in the feature space – an example is coding for object orientation in the visual system
November 7, 2004
Memory and Cognition
Page 29
Information Encoding in Neural Systems
Figure 5: Population Neuronal Codes
November 7, 2004
Memory and Cognition
Page 30
Information Encoding in Neural Systems
Temporal Binding • A single object in the real world may be encoded in the synchronous firing of neurons that code for each separate feature of the object • Multiple objects can be represented simultaneously and distinguished by the neurons representing one object firing out of phase with the neurons representing another object (phase synchronization) • Coincidence detection by a downstream neuron results in that neuron responding to higher level features e.g. red spheres
November 7, 2004
Memory and Cognition
Page 31
Information Encoding in Neural Systems
(a) Synchronous firing
(b) Coincidence Detection
Figure 6: Temporal Binding
November 7, 2004
Memory and Cognition
Page 32
Information Encoding in Neural Systems Synchronization and Oscillation • There is evidence for an increase in synchrony between cortical neurons responding to a stimulus, without necessarily any change in their average firing rates – such synchrony is seen in nearby neurons, between neurons in different cortical areas and even across hemispheres – an example is in the auditory system, where in response to a sound stimulus, neurons do not change their average firing rate, but they do fire more in synchrony with each other - this synchrony may be detected by downstream neurons, thus recognising the presence of the sound • Synchrony may coincide with, but does not depend on, oscillations in neural activity – oscillations in the gamma frequency range (20-70Hz) are seen in November 7, 2004
Memory and Cognition
Page 33
Information Encoding in Neural Systems
many cortical and subcortical regions – many other oscillation frequencies are seen in different areas and in different circumstances – theta frequency (7-12Hz) oscillations in the hippocampus are associated with exploratory behaviour in rats – such oscillations may simply be epiphenomena resulting from circuit dynamics – however, they can also act as clock signals ∗ information about spatial location is contained in the firing rates of cells in the hippocampus (place cells) ∗ But more spatial information is contained in the phase of place cell firing relative to the theta rhythm.
November 7, 2004
Memory and Cognition
Page 34
Unified Cognition What We know already? • Neurons in cortex form massive connections with other neurons forming a dense network of connectivity among themselves spanning the entire cortical section. • Brain networks are not random, they form highly specific patterns. • Different parts of brain carry out different specialized activities and thus respond to different kind of stimuli. The study of Neural Basis of Memory would be incomplete if we don’t discuss the basis of cognition. Do we have answers to following questions? • How do we have a coherent view of our experiences? • How do different neuronal population interact with each other? November 7, 2004
Memory and Cognition
Page 35
Unified Cognition
Segregation and Integration Networks in cerebral cortex possesses two kinds of structural and functional organization: Segregation: The existence of specialized neurons and brain areas, often organized into distinct neuronal populations or cortical areas. They selectively respond to specific input features and process separate features and sensory modalities. Integration: Functional integration, on the other hand, establishes statistical relationships (temporal correlations) between different and distant cell populations and cortical areas, leading to the generation of mutual information between brain regions. Thus segregated neuronal units do not operate in isolation. Any coherent perceptual and cognitive activity requires coordinated activation (functional integration) of these distributed segregated neural assemblies. November 7, 2004
Memory and Cognition
Page 36
Unified Cognition
The Binding Problem Different processes in brain at any one time will be related to a greater extent to other processes occurring in differnt regions at the same time. Now the question arises How brain keeps track of and operates on these relations between diferent processes to integrate so that a unified cognitive experience is generated?. This unknown process is given a name called the “binding problem”. Although, the mechanism of integration is still largely not known, the most plausible candidate is the formation of dynamic links mediated by synchrony over multiple frequency bands.
November 7, 2004
Memory and Cognition
Page 37
Unified Cognition
Synchronization The brain encodes information not just in firing rates of individual neurons, but also in patterns, in which group of neurons work together. The formation of these dynamice links is mediated by phase synchronisation, which refers to the relation between the temporal structures of the neural signals regardless of signal amplitude. Evidences: • Evidences from a number of researchers show that action potentials in numerous neurons, simultaneously simulated in the cortex, exhibit synchronous firing fluctuations. • Electrophysiological analyses in cats and primates have shown that the emergence of phase synchrony over widespread cortical domain correlates with the occurrence of attentive and perceptuomotor behaviours, as well as during the execution of a learning task. November 7, 2004
Memory and Cognition
Page 38
Unified Cognition
Self Organization • The system structure often appears without explicit involvement from external agent. • Brain is a nonlinear and dynamic system with large scale interaction among neurons. The couplings among neurons changes with the interaction with environment. • Stimulus play an important role in the pattern formation. • Recently, researchers are trying to view the pattern formation in brain as a self organized process. • Evidences: – Various models using self-organization have successfully reproduced patterns found in the visual cortex – Thermodynamic laws support this concept. Stimulus increases the entropy (disturbance) in neural systems. It regains its state by forming patterns. November 7, 2004
Memory and Cognition
Page 39
Summary • Neurons are the basic structural units of Memory. • Major mechanism for memory related processes is believed to be the synaptic plasticity. • Neurons generate action potential through bio-chemical reactions. This is called neural firing. This firing depends on the stimulus. • These action potentials are transmitted from one neuron to the other through a cascade of bio-chemical reactions taking place at synaptic junctions. • Long term memory formation depends on protein synthesis which are controlled by various chemicals as well as genes. • Information is encoded in the firing pulses in various ways like rate encoding, temporal encoding (interspike interval) etc. November 7, 2004
Memory and Cognition
Page 40
• Brain has both functional segregation as well as integration. That is, functions are specialised over specific brain areas while their is temporal correlation among all these specialized areas irrespective of their locations. • There is a synchrony among various neural activities which gives rise to the unified cognition. However, it is not yet properly understood. • Self Organization is considered as one of the mechanisms by which pattern formation takes place in neural systems.
November 7, 2004
Memory and Cognition
Page 41
Thank You.
November 7, 2004
Memory and Cognition
Page 42
