User:Graeme E. Smith/Local Transport Pyramidal Layer, Laminae IIIc in core and belt: How signals from the rest of the brain mix with local transport between core and belt

In 1970, Marr suggested the CODON, a 4 layer model as a method of explaining the nature of the cerebral Neocortex, In 1983 Eccles showed clearly that the CODON model did not deal well with Laminae I of the adult brain, and that there were at least 6 layers in the Isocortex that defined the operation of the cerebral Neocortex. It has been hypothesized that the extra two layers define explicit access to Marrs essential implicit content addressable memory. However in "Architectonics of the Human Telencephalic Cortex" it becomes obvious that both Marr and Eccles were victims of the staining practices of the day, and that the complexity of the neural structures do not fit as well into a 6 layer model as they thought. In this article I explore how Layer IIIc which I have labeled the local transport layer, helps the brain deal with local transport between core and belt and signals from the rest of the brain as well.

        Architectonics is the study of how the characterization of neural tissues can tell us about the nature of the architecture of the brain. Part of the background of architectonics is based on the gradual increase in the number of staining techniques that have been found for exploring the nature of the neural architecture, and mapping of those techniques to form brain atlases. In Architectonics of the Human Telencephalic Cortex ^[1] Heiko Braak shows how combining different staining systems is more informative about the nature of the Telencephalic Cortex, than using systems based on only one technique. He also notes a sub-architecture between the detailed staining studies, and the lobe based basic anatomy of the telencephalic cortex, which characterizes cortex tissues as Core, Belt, and Associative areas within the larger lobe.
        This article is about the common architectural elements between core and belt, especially in the Occipital Lobe, the Superficial Layers, and the fibers in Layer III. Of interest is the appearance of a well characterized layer IIIc in the Occipital Granulous Belt, and what it might mean to the architecture of the cerebral cortex in that area. This article is based on a relatively new model of memory function. To fully understand this model we have to reference the work of David Marr back in 1970, when he published A Theory on the Cerebral Neocortex^[2] in this article he suggested that the bottom 4 layers of the memory were a "self-classifying content addressable memory". In 1983 JC Eccles presented an article The horizontal (Tangential Fibers) System of Laminae I of the Cerebral Neocortex^[3], in which he showed that in fact the neocortex was not a 4 layer system but a 6 layer system, with organizational detail that suggested a columnar architecture.
        Work since has shown that areas that do have a smaller number of layers exist, and these so called Allocortical Tissues tend to have 3 layers rather than 4. However some question has been raised about the possibility that the 4th layer might be simply undifferentiated. Recently speculation has been made that the 4th layer or granular layer, might be related to Cortico-Thalamic inputs from the sensory system, that are rerouted via the thalamus. This hypothesis holds true in many areas of the neocortex, but may be less valid in other areas of the telencephalic cortex. If it is true, then many of the Allocortical Tissues do not have Cortico-Thalmic inputs, which seems reasonable since they probably evolved before the rerouting of the sensory data. However this model suggests that the bottom two layers, of the Mature Isocortex, laminae V and VI are relatively recent in evolution compared to three layer Allocortical Tissues. Something that is supported by the Allocortex mapping in Braak's book, which shows a great variation in the number of layers of allocortical tissue in the area around the hippocampus. This might be an indication that tissues in this area experimented with different layering schemes before settling on the six laminae scheme that seems prevalent in the mature isocortex. In Dual Mode Cortex Hypothesis ^[4] I explore the idea that the role of the cortex is to act as a memory with two interfaces one to the implicit memory system, and one to the explicit memory system, and that the main difference is the inclusion of the bottom two layers.
         If the three layer allocortical tissue hypothesis that these early tissues are content addressable, and therefore that they are probably implicit in nature is correct, then variations to the third layer will change the nature of the implicit memory in all probability. Since layer IIIc is not consistently determinable in all three layer allocortical and mature isocortical tissues, It must most probably be an alteration in the nature of implicit memory.
        The layer IIIc architectonics are therefore of interest to me, especially in areas like the boundary between the core and belt areas in the occipital cortex where the changes in Layer IIIc are abrupt and noticeable. Since my memory model has grown to include a set of assumptions about the role of the core and belt areas, I set out to find if there was any support for my role determination in the Architectonic of this boundary. Without going into detail which would be outside the scope of this article, my model includes three modes of explicit memory each associated with different types of areas within the cerebral cortex, the First mode would make use of sensory data, and result in Explicit memories, the second mode would make use of primary perception from the processing of the first mode. and the third mode would make use of secondary perception from the processing of the second mode. Hypothetically I associated mode 1 with the core areas, mode 2 with the belt areas, and mode 3 with the association areas of the cortex. If this characterization was true, then there should be a link between the core area and the belt area of the Occipital Lobe. Hypothetically in order to use the data, the transportation mechanisms would have to be post the standard areas of laminae III, and before laminae IV or later laminae. So looking closer at the architectonics of the areas, I found that they shared layer IIIc, but that the characteristics of the neurons involved were different on the different sides of the boundary. Furthermore, Fibers were found in the layer, suggesting that it might after all have a communication function.
        It should be noted that this is not sufficient proof that the layer does communicate between the two areas, merely suggestive that it does, making what I am about to say, speculative. If layer IIIc in this case is indicative of a communications link between Core and Belt, then hypothetically this link should transfer data in the direction of the belt. To support this contention, I found evidence that the pyramidal neurons associated with Layer IIIc in the Core were larger than those associated with the same layer in the Belt area. Again this is suggestive that in fact the transfer is in the required direction. It has been speculated that the size of a pyramid cell, is related to the distance it must transmit its signals. The reason it is suggestive, is that the core cells must transfer data over the distance to the belt area, but the belt neurons merely transmit between vertical layers. If the speculation on size is correct, the distinct difference in size, indicates a distinct direction to the flow of data. Given that all this speculation is correct data flows from the core to the belt via IIIc.
        Given that this is true, and it is by no means proven that it is, the data from the core is being dumped into the belt, after the normal implicit addressing layers are finished processing, and there doesn't seem to be a mechanism like that found in layer IV that telegraphs them to Laminae I, so we are left with the idea that the outputs of layer III include data from three different locations, the sensory cortico-thalamic connections in Layer IV, the Laminae I connections from other areas in the cortex, documented by J.C. Eccles, and the feed from the Core area which must be taken on par with implicit memories formed by processing the data from the other locations.
        To understand what is happening here, we need to look back at the model of what the laminae V and VI are doing. In my Dual Mode Cortex Hypothesis, I state that the role of layer 6 is to stabilize data at the Neural Group level, and that Layer 5 has a role to address data at that level. In other words the reason that the transfer happens in Layer IIIc is simply because that is the layer at which data from the core can be directly compared with data from the local implicit memory, allowing new neural groups to form that include both types of data. It has been suggested that a buffer is needed for just such a pattern to emerge, and that it's job is recognition. Here we have a bufferless mechanism that uses parallelism to the same effect.