Sample Neural Models Of Behavioral Functions Research Paper. Browse other research paper examples and check the list of research paper topics for more inspiration. If you need a religion research paper written according to all the academic standards, you can always turn to our experienced writers for help. This is how your paper can get an A! Feel free to contact our research paper writing service for professional assistance. We offer high-quality assignments for reasonable rates.
The nervous system plays a role in guiding most animal and human behavior. Extensive data has been gathered on certain aspects of behavior, including perception, learning and memory, and motor function, but theories of these behaviors are not sufficiently constrained by behavioral data alone. Research on the mechanisms of behavior can only converge to a unique solution if theories include anatomical and physiological constraints on the substrates of behavior. Extensive data has demonstrated the physiological and anatomical properties of individual neurons and large-scale brain structures. However, our understanding of speciﬁc behaviors in terms of the complex interactions within neural networks remains incomplete. The development of theoretical models and computer simulations has greatly assisted research on the neural basis of behavior.
Academic Writing, Editing, Proofreading, And Problem Solving Services
Get 10% OFF with 24START discount code
Neural models can be designed with different levels of physiological and anatomical detail. Less detailed models place less demand on computing resources, and have greater ﬂexibility for rapidly addressing a range of behavioral tasks. However, these less detailed models do not address various constraints available from physiological and anatomical data, and therefore may be less accurate in their simulation of the actual function of neural systems. This research paper will focus on more detailed models of the dynamics of individual brain regions and neural circuits involved in speciﬁc behaviors—models which draw explicitly on available data from neuroanatomy and neurophysiology. These studies all fall within the ﬁeld of computational neuroscience. At this level of analysis, considerable work has focused on speciﬁc brain regions, such as the primary visual cortex or the hippocampus. The relative focus on these particular regions arises from the more extensive experimental data available in these regions.
1. Neural Models Of Visual Cortex And Perception
Extensive anatomical and physiological research on the primary visual cortex in cat and monkey provides excellent data to constrain models addressing a range of different functional properties of primary visual cortex. This work has contributed important theoretical concepts to the empirical ﬁeld—including the concept of self-organization of feature detectors and topographic maps. Self-organization refers to a mechanism in which a population of units with random initial connectivity compete among themselves to represent salient features of the input patterns. This results from synaptic learning rules which contain some dependence upon presynaptic and postsynaptic activity, combined with some mechanism for competition between the growth of different synapses (such that growth in one set of inputs results in weakening of another set of inputs). This basic principle has been used in numerous models of the development of neurons selective to speciﬁc stimulus features. In addition, numerous models have addressed the development of the functional topography of primary visual cortex, including the mapping of adjacent locations on the retina to adjacent locations in the primary visual cortex (retinotopy), and the distribution of ocular dominance and orientation selectivity (see review in Erwin et al. 1995). Predictions from these models have led to speciﬁc experiments on the role of neural activity in cortical development, and on the speciﬁc properties of the long-term modiﬁcation of synaptic strength. Other models of primary visual cortex have addressed issues such as the mechanisms for direction selectivity of motion-sensitive neurons (Somers et al. 1995) and the physiological basis of psychophysical phenomena such as illusory contours (Grossberg and Williamson 2001).
Less data are available on the anatomical structure of regions of the cortex involved in the processing of more complex visual features. However, physiological data demonstrate stimulus selective properties of some of these regions, which have been analyzed in neural models. For example, the task of reaching for an object guided by vision cannot be done on the basis of where the object falls on the retina alone. Computing the correct reaching direction requires a complex computation involving awareness of eye direction and head direction. A neural network given input combining retinal position with eye position develops units with selectivity for speciﬁc combinations, which corresponds to unit selectivity of neurons in the parietal cortex, and provides enhanced understanding of how these representations are essential to guiding behavior (Mazzoni et al. 1991, Pouget et al. 1999).
2. Neural Models Of Hippocampus And Prefrontal Cortex In Memory
The extensive physiological, anatomical, and behavioral data on rat hippocampal formation has led to a similar focus of computational modeling on this structure. Some detailed models of hippocampal function address the network basis of physiological phenomena, without focusing on behavioral manifestations of these phenomena (Traub et al. 1992). However, numerous groups have developed detailed functional models of memory encoding and retrieval within the hippocampal formation, with explicit functional roles assigned to the dentate gyrus, region CA3, and region CA1 (Marr 1971, McNaughton and Morris 1987, Treves and Rolls 1994, O’Reilly and McClelland 1994, Hasselmo and Wyble 1997; see review in Hasselmo and McClelland 1999). These models utilize several common features which stand as important theoretical contributions to the understanding of hippocampal function.
One key idea which has emerged from this work is the idea that the overlap of the patterns of neural activity representing different inputs must be minimized in order to avoid interference between stored patterns; many models assign this function of pattern separation to the dentate gyrus. Another common principle in models of the hippocampus concerns the use of excitatory recurrent connections in region CA3 to encode different components of an episodic memory. In most of these models, associations between different elements of an episode involve activity cooccurring during a short time window in different populations of neurons in region CA3. For example, in learning an arbitrary association between two words in a paired associate memory task (e.g., dishtowel– locomotive), the co-occurrence of presynaptic activity in one population with postsynaptic activity in another population results in strengthening of the excitatory connection between the two populations. Subsequently, activation of one population by a cue word (e.g., dishtowel) causes activity to spread across the strengthened excitatory connection to activate the second population (representing the word ‘locomotive’)—allowing retrieval of the episodic association.
Modeling these associative properties of region CA3 demonstrated the dynamical requirement that the excitatory afferent input should be stronger during encoding, whereas the excitatory recurrent connections should be stronger during retrieval. Competing models have obtained these dynamics through the use of different ﬁber systems (Treves and Rolls 1994) or modulatory regulation of synaptic strength (Hasselmo and Wyble 1997). Models have also generated functional hypotheses concerning the convergence of input from region CA3 and the entorhinal cortex on region CA1. This convergence could allow retrieval from region CA3 to be compared with current sensory input and goal representations from the entorhinal cortex to determine whether the retrieval is valid on the basis of its match with the cue information. Many of the initial models of hippocampal memory function focus on static patterns of activity representing locations or individual memories, but newer models focus on encoding and retrieval of sequences of activity (Levy 1996).
Physiological data and lesion data suggest that associations initially encoded in the hippocampus gradually guide changes in semantic representations in the neocortex. Computational modeling demonstrates a functional necessity for this two-stage process (McClelland et al. 1995), illustrating that episodic memory can be readily formed with a single learning event, whereas semantic memory requires interleaved activation of multiple different episodic representations in order to prevent new information from distorting previous semantic representations.
Numerous cognitive processes require maintenance of information in working memory. For example, performance of simple delayed matching to sample tasks requires mechanisms for maintaining information during the delay period. This maintenance of information could be mediated by sustained neural activity within cortical structures, and neural models have demonstrated mechanisms for this sustained neural activity on a network level—using excitatory feedback (Lisman et al. 1998), and on a single neuron level—using self-sustaining membrane currents (Fransen et al. 1999).
Detailed physiological models have also been developed on other aspects of learning and memory function for which extensive physiological and anatomical data have become available. For example, detailed neural models of the role of the cerebellum in classical conditioning have been developed (Medina and Mauk 2000), and other neural models of conditioning are presented in Learning and Memory: Computational Models and Learning and Memory, Neural Basis of. Modeling has also elucidated mechanisms for operant conditioning, in which the activity of dopamine neurons originally triggered by reward can shift to respond to earlier stimuli which are predictive of that reward. This process provides mechanisms for selective strengthening of connections mediating goal-directed behavior (Schultz et al. 1997).
3. Models of Small Neural Circuits
Another approach to understanding the neural basis of behavior focuses on simpler neural circuits in nonmammalian preparations. Nonmammalian preparations often have smaller numbers of neurons than mammalian structures, and sometimes these neurons are more accessible to detailed physiological recording than those in mammalian preparations. The smaller number of neurons allows the creation of models incorporating speciﬁc features of individual neurons, and analysis of the dynamical interactions of these modeled neurons (Calabrese and De Schutter 1992). Nonmammalian preparations have numerous neural specializations that may result in fundamental differences between the neural mechanisms in these animals compared to mammals. However, important convergent properties have been described in other species, including principles of self-regulation, in which individual neurons adapt their cellular properties so that they maintain a similar level of activity, even after the circuit interactions have been perturbed (Golowasch et al. 1999). With regard to motor function, these smaller scale and more extensively described neural circuits provide the opportunity for more detailed analysis of the neural circuits underlying the generation of speciﬁc motor patterns (Cohen et al. 1992, Calabrese and De Schutter 1992).
4. Models Which Include Ongoing Interaction With The Environment
The models described above provide effective mappings of data on input parameters to speciﬁc behavioral output, but they do not include an alteration of the environmental input which depends upon the current output of the model. This can only be obtained with explicit robotic implementations, or with representations of a virtual organism moving through a virtual environment. Ultimately, accurate representation of the temporally continuous interaction with the environment during behavior will require models with environments inﬂuenced by the model output.
A few examples have explicitly demonstrated the ongoing interaction of a model with its environment. For example, modeling of spatial navigation requires an ongoing interaction of the model with the environment, as the movement of an animal through the environment under the control of the model will result in constant updates of its position. Several simulations have demonstrated how the place cell representations formed in the hippocampal formation can be used to guide movement through an environment (Sharp et al. 1996, Burgess et al. 1997, Redish and Touretzky 1998). These models provide excellent examples of the necessity for a constant interactive updating of parameters including the sensory input from the environment, and the proprioceptive knowledge of current heading (head direction). Neural models also demonstrate mechanisms for maintenance of this proprioceptive information, using attractor dynamics within regions encoding the head direction (Sharp et al. 1996).
The production of complex sounds or movements requires constant monitoring of feedback from the sensory periphery and from the environment in order to monitor the appropriate process of production. Sophisticated models have moved beyond a simple feed-forward output and have addressed the continuous dynamical interaction of motor output and sensory input (Troyer and Doupe 2000).
In the realm of experimental data, there continues to be a gap between the knowledge of anatomical and physiological properties in neural systems and the knowledge of how behavior could be mediated at a neural level. Neural models of behavior will provide an essential theoretical link between these different levels of analysis—similar to the role of theoretical physics in understanding the properties of physical systems. Neural models of behavior have advanced dramatically in recent years, both in their level of anatomical and physiological detail, and in their capacity to ﬂexibly address a broader scope of behavioral data. Computational models have already provided important theoretical concepts to the empirical work. As empirical techniques become more sophisticated for analyzing complex systems, computational models will become increasingly important in understanding the theoretical principles involved in the neural basis of behavior.
- Burgess N, Donnett J G, Jeffery K J, O’Keefe J 1997 Robotic and neuronal simulation of the hippocampus and rat navigation. Philosophical Transactions of the Royal Society of London Series B—Biological Sciences 352(1360): 1535–43
- Calabrese R L, De Schutter E 1992 Motor-pattern-generating networks in invertebrates: Modeling our way toward under-standing. Trends in Neuroscience 15(11): 439–45
- Cohen A H, Ermentrout G B, Kiemel T, Kopell N, Sigvardt K A, Williams T L 1992 Modelling of intersegmental coordination in the lamprey central pattern generator for locomotion. Trends in Neuroscience 15(11): 434–8
- Erwin E, Obermayer K, Schulten K 1995 Models of orientation and ocular dominance columns in the visual cortex: A critical comparison. Neural Computation 7(3): 425–68
- Fransen E, Wallenstein G V, Alonso A, Dickson C T, Hasselmo M E 1999 A biophysical simulation of intrinsic and network properties of entorhinal cortex. Neuro-computing 26 –27: 375– 80
- Golowasch J, Casey M, Abbott L F, Marder E 1999 Network stability from activity-dependent regulation of neuronal conductances. Neural Computation 11: 1079–96
- Grossberg S, Williamson J R 2001 A neural model of how horizontal and interlaminar connections of visual cortex develop into adult circuits that carry out perceptual grouping and learning. Cerebral Cortex 11(1): 37–58
- Hasselmo M E, McClelland J L 1999 Neural models of memory. Current Opinion in Neurobiology 9: 184–8
- Hasselmo M E, Wyble B P 1997 Simulation of the effects of scopolamine on free recall and recognition in a network model of the hippocampus. Behavioral Brain Research 89: 1–34
- Levy W B 1996 A sequence predicting CA3 is a ﬂexible associator that learns and uses context to solve hippocampal-like tasks. Hippocampus 6: 579–90
- Lisman J E, Fellous J M, Wang X J 1998 A role for NMDA receptor channels in working memory. Nature Neuroscience 1(4): 273–5
- Mazzoni P, Andersen R A, Jordan M I 1991 A more biologically plausible learning rule than back propagation applied to a network model of cortical area 7a. Cerebral Cortex 1(4): 293–307
- McClelland J L, McNaughton B L, O’Reilly R C 1995 Why there are complementary learning systems in the hippocampus and neocortex: Insights from the successes and failures of connectionist models of learning and memory. Psychological Review 102: 419–57
- McNaughton B L, Morris R G M 1987 Hippocampal synaptic enhancement and information storage within a distributed memory system. Trends in Neuroscience 10: 408–15
- Medina J F, Mauk M D 2000 Computer simulation of cerebellar information processing. Nature Neuroscience 3: 1205–11
- O’Reilly R C, McClelland J L 1994 Hippocampal conjunctive encoding, storage, and recall: Avoiding a trade-off. Hippo-campus 4(6): 661–82
- Pouget A, Deneve S, Sejnowski T J 1999 Frames of reference in hemineglect: A computational approach. Progress in Brain Research 121: 81–97
- Redish A D, Touretzky D S 1998 The role of the hippocampus in solving the Morris water maze. Neural Computation 10(1): 73–111
- Schultz W, Dayan P, Montague P R 1997 A neural substrate of prediction and reward. Science 275: 1593–9
- Sharp P E, Blair H T, Brown M 1996 Neural network modeling of the hippocampal formation spatial signals and their possible role in navigation: A modular approach. Hippocampus 6(6): 720–34
- Somers D C, Nelson S B, Sur M 1995 An emergent model of orientation selectivity in cat visual cortical simple cells. Journal of Neuroscience 15(8): 5448–65
- Traub R D, Miles R, Buzsaki G 1992 Computer simulation of carbachol-driven rhythmic population oscillations in the CA3 region of the in vitro rat hippocampus. Journal of Physiology (London) 451: 653–72
- Treves A, Rolls E T 1994 Computational analysis of the role of the hippocampus in memory. Hippocampus 4(3): 374–91
- Troyer T W, Doupe A J 2000 An associational model of birdsong sensorimotor learning I. Efference copy and the learning of song syllables. Journal of Neurophysiology 84(3): 1204–23