Difference between revisions of "BiologicalModel"

From aHuman Wiki
Jump to: navigation, search
Line 13: Line 13:
 
* L2: Split into items; connections are classified, function is well-defined by connections
 
* L2: Split into items; connections are classified, function is well-defined by connections
 
** '''driver''' - single primary source input
 
** '''driver''' - single primary source input
** '''corrector''' - single secondary input to perform region function (optional)
+
** '''corrector''' - single secondary input to perform region function (optional, if region only bypasses signal, few cases only)
 
** '''secondary''' - multiple secondary input to affect region function (optional)
 
** '''secondary''' - multiple secondary input to affect region function (optional)
** '''modulator''' - multiple neuromodulator input affect region operating mode (optional)
+
** '''modulator''' - multiple neuromodulator input to affect region operating mode (optional)
 
* L3: Properties are specified and domain-aligned; region, all its inbound and outbound connections are covered by references
 
* L3: Properties are specified and domain-aligned; region, all its inbound and outbound connections are covered by references
  

Revision as of 14:49, 28 April 2019

Biological Model

Home -> BiologicalLifeResearch -> BiologicalModel


Model Quality

Region Quality Levels

  • L0: Identified; all required properties are specified
  • L1: Connections are representative enough, can include possible replicas (not a true connection, registered by transitive signal)
  • L2: Split into items; connections are classified, function is well-defined by connections
    • driver - single primary source input
    • corrector - single secondary input to perform region function (optional, if region only bypasses signal, few cases only)
    • secondary - multiple secondary input to affect region function (optional)
    • modulator - multiple neuromodulator input to affect region operating mode (optional)
  • L3: Properties are specified and domain-aligned; region, all its inbound and outbound connections are covered by references

Circuit Quality Levels

  • L0: Defined (probably just a group of regions, having poor coupling)
  • L1: Has well-defined name; contains no more that 10 regions, all target regions are at least L2
  • L2: Circuit is well-defined, stable, all circuit regions are at least L2

Model Items

Modality

  • primary feature of mind elements is a modality (M)
    • all muscles are the same modality, all elements of specific sense (vision, audition, tactile, taste, vestibular) are the same modality
    • modality can be derived from other modalities, e.g. spinocerebellum has a modality, merged from muscle command and muscle proprioception
    • modality is like a color in colored directed graph, where every edge (G) and node (N) has specific color
    • modality of node is always equal to the modality of all output edges, edge modality (M(g))
    • modality defines limited and specific modality field (Fm), e.g. body skin produces surface with final resolution, number of points is defined by density of tactile receptors
    • single neuron always has a specific modality and corresponds to modality field point (Pm[k])
    • one modality can be mapped to another modality, e.g. tactile modality from body skin can be easily mapped to pain modality from skin, or to body muscles via more discrete "body part"

Elements

  • mind element (E)
    • element is a name, designating set of mind neurons, that can be decomposed to final neurons
    • one element can include another elements, and in general is non-exclusive grouping over mind neurons, e.g. overall mind is also an element
    • mind can be split into smallest elements, which have specific modality, and mapped to specific part of modality field, modality subfield (Fm(e))
    • can be too specific, e.g. part of motor cortex, corresponding to thumb finger of left hand
    • can be too generic, e.g. extended amygdala complex, which includes elements located quite far from each other, or elements having different modalities
    • specific element has specific modality
    • generic element has no specific modality
      • e.g. thalamus has no specific modality because its nuclei relay signals of different nature (LGN-visual, MGB-audio, VPL-tactile, VPI-pain, VA-motor command ...)
    • edges of output projections of the element have the same modality subfield, edge subfield (Fm(e)=Fm(g), where e has modality and g is output edge of e)

Components

  • mind elementary component is defined by:
    • creating strong hierarchy of elements, using known names, and selecting specific cut of this hierarchy as flat list of mind components
    • joining into component most of neighbor elements, located together, close to each other
      • e.g. VPL is a component while basal ganglia complex is not a component
    • using only elements having the same set of input modalities
      • e.g. VPL receives sensory information from CU,GR nuclei which are nuclei of the same skin tactile modality, having lower and upper body subfields
      • e.g. every part of spinocerebellum has mossy fiber inputs of muscle command modality, and climbing fiber inputs of muscle proprioception modality
  • every mind component produces specific component modality
    • can be one of afferent modalities or new generated modality
    • component is mapped to specific modality subfield
    • in case of modality derived from source, component subfield is equal to the union of afferent (input) edge subfields
  • mind has no component, which can be replaced by direct connection from afferent source to efferent destination, but it is always one of:
    • superposition of two or more major inputs to produce new value, can produce new modality (A + B -> C)
    • place to affect one major input by secondary minor inputs, stay with major input modality (A* + B -> C*)
    • send source to two or more distinct targets (A -> B,C)