|
Hermes Hybrid
Brain Model
The Hermes hybrid brain model is a new
research and development project with the goal of implementing a
working software model of biological brain function based upon
current research into the role of astrocytes in brain function and
consciousness. This model is based upon "The Astrocentric
Hypothesis: proposed role of astrocytes in consciousness and memory
formation" which was authored by Dr. James M. Robertson, and
published in Journal of Physiology - Paris 96 (2002).
To date, all of the Artificial Neural Network (ANN) implementations, be they software or hardware, have been extremely oversimplified versions of neural functioning, and have completely ignored the function of the majority of cell types in the brain. Dr. Robertson proposes that there is an anatomical entity in the brain that binds parcelated sensory information in a unified manner. Protoplasmic astrocytes in the cerebral cortex have the necessary physical, chemical, and structural elements to integrate sensory information.
Our initial working hypothesis is that neural structures of the brain provide temporal, contextual, and relational structure to the information and cognitive agencies expressed within the astrocytic syncytium. This type of model would explain and overcome the ANN shortcomings so far experienced.
The eventual desired goal is to prove this
concept via software model, so that we can proceed with developing
hybrid digital/analog devices in a larger scale experiment.
The potential benefit to such a hardware based system in robotic and
cognitive applications can not be overstated. We hope to
develop hardware prototypes in conjunction with our software models.
Important References:
J.M. Robertson, The Astrocentric Hypothesis: proposed role of astrocytes in consciousness and memory formation, Journal of Physiology - Paris 96 (3-4) (2002) pp. 251-255.
Abstract
Consciousness is self-awareness. This process is closely associated with attention and working memory, a special form of short-term memory, which is vital when solving explicit task. Edelman has equated consciousness as the "remembered present" to highlight the importance of this form of memory (G.M. Edelman, Bright Air, Brilliant Fire, Basic Books, New York, 1992). The majority of other memories are recollections of past events that are encoded, stored, and brought back into consciousness if appropriate for solving new problems. Encoding prior experiences into memories is based on the salience of each event (A.R. Damasio, Descartes' Error, G.P. Putnam's Sons, New York, 1994; G.M. Edelman, Bright Air, Brilliant Fire, Basic Books, New York, 1992). It is proposed that protoplasmic astrocytes bind attended sensory information into consciousness and store encoded memories. This conclusion is supported by research conducted by gliobiologist over the past 15 years.
Link to paper:
http://www.ncbi.nlm.nih.gov/entrez/queryd.fcgi?cmd=Retrieve&db=PubMed&list_uids=12445903&dopt=Abstract
Perea G, Communication between astrocytes and neurons: a complex language. J Physiol Paris. 2002 Apr-Jun;96(3-4):199-207.
Abstract
In recent years, accumulating evidence suggests the existence of bidirectional communication between astrocytes and neurons, indicating an important active role of astrocytes in the physiology of the nervous system. As a consequence of this evidence, a new concept of the synaptic physiology "the tripartite synapse" has been proposed, in which the synapse is formed by three functional elements, i.e. the pre- and postsynaptic elements and the surrounding astrocytes. In the present article we review and discuss the current knowledge on the cellular mechanisms and physiological properties of this communication that displays highly complex characteristics. We are beginning to realize that the communication between astrocytes and neurons uses a quite complex language.
Link to paper:
Zonta M, Calcium oscillations encoding neuron-to-astrocyte communication. J Physiol Paris. 2002 Apr-Jun;96(3-4):193-8.
Abstract
The observation that the excitatory neurotransmitter glutamate released from presynaptic terminals can activate, beside the post-synaptic neuron, the glial cell astrocyte, stimulated glial cell research like no other event since the recognition in the 1980s that astrocytes can express on their membrane many receptors for classical neurotransmitters. The properties and the functional role(s) of such a neuron-to-astrocyte signaling have now become the focus of intense research in neurobiology. Indeed, a growing body of evidence has recently highlighted the ability of astrocytes to work as sophisticated detectors of synaptic activity: by changing the frequency of [Ca2+]i oscillations evoked by the synaptic release of glutamate, these cells display the remarkable capacity to discriminate between different levels and patterns of synaptic activity. Furthermore, the observation that astrocytes increase the frequency of [Ca2+]i oscillations in response to repetitive episodes of high neuronal activity challenges the common concept that memory function in the brain is an exclusive property of neuronal cells. Glutamate-mediated [Ca2+]i elevations can also trigger in astrocytes the release of glutamate that can ultimately affect neuronal transmission. Given the wide role played by glutamate in brain physiology, our view on how the brain operates needs now to be revised taking into account the bi-directional, glutamatergic communication between neurons and astrocytes.
Link to paper:
Alejandro F Frangi, Gap Junctions for Engineers: A review
Link to pdf:
http://petrus.upc.es/~wwwdib/bioz/caract_tissues/ files/grepdoc.pdf
Contact info@artificialingenuity.com
Copyright © 2005 Artificial Ingenuity, LLC
Last modified: June 11, 2005
Initial design by Webinizer, LLC