3/14/2023 0 Comments Neural processesIf an input feature is defined as the sum of graded potentials that arise at the soma due to a specific presynaptic neuron but with some possible time variation across the many input synapses from this specific neuron, then it is clear that an input feature could arise due to either a binary or continuous variable. So it would be reasonable to believe that each neuron receives many synaptic inputs from the same presynaptic neuron. In real neurons, an axon may have terminal branches to thousands of synapses, 9 and many of these could terminate on the same target neuron. Each weighted feature effect arises directly either through synaptic inputs or through their associated interaction and nonlinear effects as determined by neural dynamics. 2.2 are akin to graded potential effects at the soma in a real neuron where the weightings of step 1 are prior weights. The weighted feature effects in the RELR model that are summed in step 2 in Fig. Figure 2.2 describes all steps in the basic RELR computational process that are seen in both Implicit and Explicit RELR learning. So, as a neural computational model, RELR is a mechanism for how binary on–off spiking responses would be learned as a function of these many synaptic input signals and their corresponding interaction and nonlinear effects. In real neurons, this needs to be done so that reliable and stable learning can occur quickly across the large number of synaptic inputs and associated interaction and nonlinear effects as would be measured at the point where the soma meets the axon, which is the initial zone within the neuron where axonal spiking occurs. RELR is a mechanism for the computation of memory weights that ultimately determine binary on–off spiking responses. (For color version of this figure, the reader is referred to the online version of this book.) Gaps where information jumps to neighboring neurons are called synapses which are usually bridged at axonal terminals through chemical messengers called neurotransmitters, although electrical tight junction synapses also exist. 8 Neural electric information flows from dendrites to cell bodies to axons. The cell bodies are also obvious as the oval-shaped structures in the middle, and axons are seen exiting these cell bodies. The dendrites of the Purkinje cells are easily seen at the top as abundant tree-like structures. Purkinje cells are exemplified by (A) and granule cells by (B). Drawing of neurons in pigeon cerebellum by Santiago Ramón y Cajal, 1899 Instituto Cajal, Madrid, Spain.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |