Script – The Hippocampal Region

================

— POSITION —

================

[ HIPPOCAMPAL REGION ]

THE FLOW OF INFORMATION

We have been following the flow of information from the sensory nervous system, to the brain stem, to the brain, to the cortex, it’s neurons, and columns, modules, regions, and networks.

We’ve seen that cortical layers 1-3 calculate WHAT, and layers 4-6 calculate WHERE.

We’ve seen that sequences of stimulation predict fragments, predict objects, predict models, predict entities.

We’ve seen that the dorsal (boundaries) and ventral (objects) visual pathways separate prediction of scene and object  information.

We’ve seen how our orientation assists columns in the matching or resolution of different senses, objects, and scenes.

And we’ve seen at least two sources of that information: the physical, and the visual, … and that the visual … which constitutes is half of our nervous system, splits information into object and scene components. I suppose we should talk about that at some point: that a wall of trees is a single spatial object, and a tree blocking your path is  a single object. So it might be more correct to say that we disambiguate actionable from inactionable things, from visible space – and we just say objects and scenes or objects and borders as if they are different things rather than different degrees of resolution.

But, that said, what happens with that information? Where does that information go, how is it organized, what is it used for, and how is that orientation calculated?

We need a little more framework to do that.

So, LET’S GO BACK TO GEOMETRY FOR A MOMENT – and KEEP THE PARALLEL GOING

If we were going to produce a 3d video game, we would need the following variables:

WORLD MODEL –

WORLD

LOCATIONS SCENES, PATHS  ROOMS, (CONTEXTS)

BOUNDARIES AND DISTANCES (LIMITS)

SPACES(GRID) OPEN TO ACTION (OR NOT) (POSSIBILITIES)

PLACE THOSE SPACES

OBJECTS IN IT

OBJECTS (SPACE OCCUPIERS) OPPORTUNITIES OR INHIBITORS

OBJECTS IN MOTION (CHANGE)

OURSELVES

DIRECTION OF TRAVEL IN ENVIRONMENT, (KEYBOARD W-forward S-backward)

DIRECTION OF TURN IN ENVIRONEMNT, (KEYBOARD A-left D-right)

ORIENTATION IN INVIRONMENT (rotation from landmark)

LIMB MOVEMENTS (CHANGES)

HEAD DIRECTION (MOUSE LOOK)

EYE DIRECTION  (YOUR EYES IN THAT MOUSE LOOK)

MOVEMENT

DISTANCE TRAVELLED (ESTIMATED)

SPEED (VELOCITY IN TIME USING LIMB MOVEMENTS)

TIME (RELATIVE CHANGE)

SYSTEM OF MEASUREMENT (ACTION)

body (action),

emotional (incentives, values) )

mind (prediction, planning, and choice),

RECALL – EVENTS

INDEX

EPISODE (EVENT)(I prefer event memory myself)

MEMORY

(what where when who how why)

And then let’s move from an event to a narrative model (sequence of events)

  NARRATIVE OR DESCRIPTIVE MODEL (CAUSAL)

WHAT – Object

WHERE – Space and Location

WHEN – very iffy, by association

WHO – face, object

HOW – Sequence, causality, imitation of action

WHY – incentives: imitation of sympathy and empathy

THE IMAGINARY MODEL (POSSIBILITIES)

We’ll get there.

THE THREE GENERATIONS OF THE BRAIN

As we’ve discussed already, there are sort of three generations of the brain,

– the reptillian brain, which consists largely of autonomic functions

– the limbic brain, wrapped around it like a glove.

– the and the neocortex wrapped around the limbic brain like a big puffy glove.

In this section we’re going to discuss some of the functions of the limbic system, and in doing so learn where all that information we’ve disambiguated goes.

IN THIS SECTION

This is the Hippocampal region. We’ll cover it’s connection with the neocortex, and  its role in organiztion and integration into a coherent experience, and touch on imagination, concentration, and memory of those integrated real and imagined episodes.

Now, caveat: I’ve tried producing and delivering this lecture four times now, and it’s very easy to get lost in the biology and lose sight of the function. So, my goal, our goal, is a little less resolution about how each organ or region of the organ, functions, and a little more resolution on what information it’s producing.

Hopefully I’ve found the right balance. We’ll see.

HIPPOCAMPAL FORMATION

[ SHOW DIAGRAM OF FORMATION – YELLOW GREEN ORANGE]

The Hippocampal formation consists of

The

Parahippocample  Boundaries in relation to you (where)

Perirhinal, and  Objects in Relation to you (what)

Entorhinal,  You. (you)

cortices.

the Hippocampus proper, (location)

Consisting of

The Dentate gyrus  (input)

The Cornu Ammonis  Or C.A. for short

The subiculum,  (output)

And the output of the hippocampus

The Fimbria

The Fornix (hemispheric coordination)

The Mammilary Bodies  (arguably part of the thalamus, seem to affect spatial memory, head direction)

The Anterior (front) Nucleus of the Thalamus (Attention)

We aren’t going to cover:

The Hypothalamus

(links nervous system to the endocrine system via the pituitary gland, controls body temperature, hunger, important aspects of parenting and attachment behaviours, thirst, fatigue, sleep, and circadian rhythms.)

The Amygdala

(emotion: fear, anxiety, and aggression, emotional weight of memories, sex differences in behavior, and sex differences in political orientation (esp sensitivity to disgust, threat.))

THe Cingulate Gyrus ( complicated )

(receives inputs from the thalamus and the neocortex, and projects to the entorhinal cortex via the cingulum.)

The Cingulum (pain)

(CUT THIS)

OVERVIEW OF HIPPOCAMPAL FORMATION INFORMATION PROCESSING

[ DIAGRAM oF FORMATION (YELLOW) ]

The hippocampal Network: The hippocampus forms a largely uni-directional network, with input from the Entorhinal Cortex (EC) that forms connections with the Dentate Gyrus (DG) and CA3 pyramidal neurons via the Perforant Path (PP – split into lateral and medial).

CA3 neurons also receive input from the DG via the Mossy Fibres (MF). They send axons to CA1 pyramidal cells via the Schaffer Collateral Pathway (SC), as well as to CA1 cells in the contralateral hippocampus via the Associational Commisural (AC) Pathway. CA1 neurons also receive inputs direct from the Perforant Path and send axons to the Subiculum (Sb). These neurons in turn send the main hippocampal output back to the EC, forming a loop.

THE VISUAL PATHWAY FEEDS THE FORMATION

The first neocortical area to process the visual information is the primary visual area (V1), the lowest area in the visual neocortical hierarchy; V1 processes visual information in its rawest, most basic form, after which it may be processed further up the hierarchy by higher association areas called V2, V3, V4, V5/MT, and the inferotemporal cortex ( or IT for short).

The Inferotemporal cortex or IT is the last and highest area in the neocortical hierarchy.

The hierarchy isn’t limited to vision; there are also neocortical hierarchies all the other senses such as touch, hearing, and taste are also organized into hierarchies – and they integrate with each other.

  That entire visual neocortical hierarchy converges onto the perirhinal and parahippocampal cortices, which in turn converge onto the entorhinal cortex; ultimately, the whole neocortical hierarchy converges onto the hippocampus.

So do – all of the senses, really. We’re just using vision because it’s dominant, and because I feel by using touch in our previous lectures, we’ve defended against overemphasis on vision.

From there the information continues through the Fornix, back into the thalamus, and then back into the neocortex.

That’s our main circuit.

Although there is a lot more going on than that, because all these regions transfer information back and forth like eddies in a stream, in a process called back-propagation that feeds information back up the information stream.

All you need to grasp is that information is collect by our senses, associated and disambiguated in our cortex, and that set of information flows downward and inward where we start to integrate those pieces into a spatial composition relative to our bodies.

Information is fed from the IT, into the  THE ENTO-RHINAL, PERI-RHINAL, AND PARA-HIPPOCAMPLE CORTICES along with information from the Cingulate Gyrus and the Cingulum

THE CINGULATE GYRUS

The cingulate gyrus has four regions, and that each provides a unique contribution to brain functions. I’ll call them 1, 2, 3, and 4 to spare you their names.

(Cut)

(These regions and subregions are the subgenual and pregenual anterior cingulate cortex (sACC and pACC), the anterior and posterior midcingulate cortex (aMCC and pMCC), the dorsal and ventral posterior cingulate cortex (dPCC and vPCC), and the retrosplenial cortex (RSC). )

The Cingulate gyrus has three roles in pain processing.

1. The first is involved in unpleasant experiences and directly drives autonomic outputs.

2. Second, the is involved in fear, prediction of negative consequences and avoidance behaviours throu the motor area.

3. Third, mediates the orientation of the body in space through the caudal cingulate motor area.

Lastly, pain stimuli reduce activity and, subsequently, activity in a subregion that normally evaluates the self-relevance of incoming visual sensations.

I suspect this region allows us to rapidly seize attention and react to pain draw attention to pain, regardless of whatever else is consuming our attention.

THE CINGULUM

the cingulum is a collection of white matter, with fibers projecting from the cingulate gyrus to the entorhinal cortex, connecting each area along the way, and its function appears to be the correction of error from pain to mistakes to emotions. Despite being one of the first brain structures identified, the research is pretty limited, or rather the quality of research is rather limited. Problems with the front or anterior lead to apathy and depression, schizophrenia is implicated. Damage to the back or posterior can include interference with attention, visual and spatial skills, working memory and general memory.

One bit I find interesting, is that animals that eat their young don’t have this behavioral inhibitor.

THE PARAHIPPOCAMPAL CORTEX

the parahippocampal processes geometric information about environment and  landmarks > continuing the dorsal computation. we use these landmarks to orient ourselves in a space and location.

So, NOW WE HAVE LANDMARKS – reference points.

THE PERIRHINAL CORTEX

The perirhinal cortex processes representation of complex objects, continuing the ventral computation.

Since both PARAHIPPOCAMPAL PERIRHINAL cortexes process complex information, these two cortices lie near the top of the visual neocortical hierarchy, and they physically are situated immediately above IT;

So now we have OBJECTS.

THE ENTO-RHINAL CORTEX

The ento-rhinal cortex represents us, ourselves, in relation to environemnt and objects.

It’s a region of neocortex, and is part of the hippocampal region,

It’s a strip of tissue running along the back edge of the  brain from the ventral to the dorsal sides.

It’s the richest set of association connections of any cerebral cortical region.

and  acts as the primary interface between hippocampus and neocortex.

The entorhinal cortex differs from the hippocampus proper in that it consists of six layers (L1-6) of minicolumns like the rest of the neocortex.

[DIAGRAM]

as you can see in this diagram the six layers of the ento-rhinal cortex consolidate information just as did the cortical columns, modules, and regions, and I’m not confident in this, but it looks to me like the relationship is inverted, with what and where reversing layers.

Different regions project into different levels of the hippocampus, where object and positional information is incrementally combined.

And it’s reciprocally connected to the perirhinal and parahippocampal cortices so that two functions are served: every stream is constantly calculating differences with the past moment of the stream, detecting change without needing to ‘store’ information to compare, and secondly to keep alive the network that’s producing the stream of information, so that memories are reinforced as long as they’re being used.

Now remember how each column recieved some sort of positional information? Well, The EHC is where we begin to see the rather miracuous way we create that information.

In 2005, researchers discovered that entorhinal cortex contains a neural map of the spatial environment.

These maps are produced by Neurons of the MEDIAL entorhinal cortex (MEC). They store multiple “place fields” that are arranged in a hexagonal grid – and so they’re called “grid cells”.

The medial entorhinal cortex mainly supports processing of space, and the lateral entorhinal cortex supports the processing of time.

And together the neurons process general information such as your direction of movement in the environment.

So, the EC encodes general properties about your CURRENT CONTEXTS, and The hippocampal neurons, by contrast, usually encode information about specific places. So EC current context is used by hippocampus to create unique representations from combinations of these measurements in a PLACE CONTEXT.  These places are the primary indexes that we organize our memories by.

So now we have BOUNDARIES, SCENES, OBJECTS, added SPACE AND TIME

SPEED

The same group of researchers have found speed cells in the medial entorhinal cortex. The speed of movement is translated from pro-prio-ceptive (movement of position of the body – meaning limb movement) information and is represented as firing rates in these cells. The cells fire in correlation to future speed of the animal.

Firing speed cells in response to the movement of the animal provides instantaneous running speed to the Grid cell. The grid cell in turn uses this information along with the head direction in order to calculate the location of the animal in the cognitive map.

DIRECTION OF TURN

Recording of neurons in humans playing video games found path cells in the EC, the activity of which indicates whether a person is taking a clockwise or counterclockwise path. These EC “direction” path cells show this directional activity irrespective of the location of where a person experiences themselves, which contrasts them to place cells in the hippocampus, which are activated by specific locations.

EYE MOVEMENT AND HEARING

The EC is also responsible integration of eye-tracing, sound, and smell into the information stream. Helping us sense by integration which direction and distance something in motion is from us.

Just for the record, it seems like control of eye movement, that allows us to track moving objects is stored in mutliple places rather than one. I wont go into it here, but we have a visual field in the frontal cortex just in front of the motor cortex. We have visual tracking here in the entorhinal cortex, and there is other tracking information in the brain stem, the thalamus, and probably elsewhere.

I learned quite a bit about the depth of our visual dominance producing this series of presentations. A dog may live in a world of smells, but we live in a world dominated by vision, that consumes most of the processing functionality of our nervous system – and it’s very expensive processing.

WORLD MODEL –

+PlACE

+BOUNDARIES AND DISTANCES

+SPACES(GRID)

+OBJECTS (SPACE OCCUPIERS)

OBJECTS IN MOTION

DISTANCE TRAVELLED (ESTIMATED)

+SPEED,

+TIME

+DIRECTION OF TRAVEL IN ENVIRONMENT,

+DIRECTION OF TURN IN ENVIRONEMNT

ORIENTATION IN INVIRONMENT

+LIMB MOVEMENT

HEAD DIRECTION

+EYE DIRECTION

FIELD OF VIEW

+SYSTEM OF MEASUREMENT (ACTION)

HIPPOCAMPUS AND PROJECTIONS

The hippocampus consists of a pair of seahorse-shaped cortical tissues deep in the temporal lobe.

While we talk about it as a separate organ, it’s really just an infolding of cerebral cortex – and a mixture of the larger, more recent neocortex plus the smaller, more ancient allocortex. A strongly interconnected brige between them.

It’s similar to the thalamus in the sense that the thalamus is more like another layer of the neocortex for short close transmission and very little if any neural computation, and the thalamus is connected to the cortical layers and their columns by longer transmissions to isolated mini supercomputers (the columns). So in this sense the hippocampus, neocortex and thalamus are more like layers of the same organ.

However, the hippocampus and neocortex structurally differ; the hippocampus consists of several three-layered structurally distinct subdivisions that don’t have minicolumns, where the neocortex consists of numerous six-layered, structurally similar minicolumns arranged into a sequential hierarchy of processing regions

SUBREGIONS

The Hippocamus subregions include an input region called the dentate gyrus,

an auto-association region called the Cornu Ammonis or CA for short, and an

output region called the subiculum. Both the CA and Subiculum are divided into subsections.

I want to make it stick in  your memory that again, that relative position and object or boundary are computed together, and this information has to be integrated teogether with information about your physical body, as well as your physical body in relation to some path or other.

Information runs through the dentate gyrus, then the Cornu Ammonis or CA areas, then into the subiculum where it’s output. SO think of this as the input, auto-association or composition function, and the output function.

(a) DENTATE GYRUS. INPUT AREA

The dentate gyrus is a tightly packed layer of small granule neurons wrapped around the end of the hippocampus. The dentate gyrus is the main INPUT region of the hippocampus formation and receives most of its information from the entorhinal cortex, but also from the thalamus and amygdala, among others.

(b) CORNU AMMONIS, or CA 1,2,3,4. – The Processing Area

Next come a series of Cornu Ammonis areas: first CA4 (which underlies the dentate gyrus), then CA3, then a very small zone called CA2, then CA1. The CA areas are all filled with densely packed Pyramidal cells similar to those found in the neocortex.

The CA3 neurons create autoassociations that organize episodic memories by creating associations between spatial and reward representations, that can be recalled by any fragment of that set of associations. Reward representations means value judgments, the same way saliency in the cortex means compositional value.  The CA3 area composes episodes out of a multitude of fragments – what we think of as memories.

(b) THE SUBICULUM . OUTPUT

The subiculum is the main OUTPUT region of the hippocampus formation.

It has four regions we’ll discuss in a moment.

It consists of densly packed pyramidal neurons.

It receives input from CA1 and entorhinal cortical layer III – the cortical output layer for predictions.

It projects mainly into the entorhinal cortex (you), but the pyramidal neurons send projections everywhere, and not that you’ll need to know thes but I’ll liste them: to the nucleus accumbens, septal nuclei, prefrontal cortex (working memory), lateral hypothalamus (recursion), nucleus reuniens, mammillary nuclei, entorhinal cortex  (you) and amygdala ( survival emotions of fear, anger pleasure), and these emotions either amplify and improve memory or depress and avoid rehearsal into memories.

The pyramidal neurons use two modes of output: bursting and single spiking. It’s possible that the transitions between these two modes is involved in routing information out of the hippocampus.

I’ve read all the literature on it that I can find but at present we can only say it’s not just relaying, but calculating some sort of spatial information, and that this spiking and bursting neurons – which are located at different distances from the output.

So input, integration, association, episode formation, and forwarding information.

THE SUBICULUM OR SUBICULUM COMPLEX

Now, lets move on to the regions of the Subiculum and see what it’s calculating.

There are Four component areas to describe: parasubiculum (adjacent to the parahippocampal gyrus), presubiculum, postsubiculum, and prosubiculum.

Prosubiculum – I DON”T KNOW

Prosubiculum refers to a region located between the CA1 region of the hippocampus and the subiculum, and distinguished by higher cell density and smaller cell sizes. Its in apes and humans but apparently not in rates. I am not clear on the function of the prosubiculum. There is some historical dispute over whether it’s a separate function or not, but the most recent research I can find suggests its distinctly strutured, so I am going to stick with it’s a translatoin layer until I know better.

Presubiculum – SPATIAL MEMORY – BOUNDARIES

The presubiculum serves to provide cortical input to the entorhinal-hippocampal spatial/memory system.  My understanding is that boundary cells are in both the pre and parasubiculum, and this is wherewe we organize produce boundaries into a coherent set.

Parasubiculum – DIRECTION AND DISTANCE – MEASUREMENT OF ACTIONS

The parasubiculum contains grid cells, which are neurons responsive to movements in particular directions over particular distances.

Postsubiculum HEAD DIRECTION

The dorsal part of the presubiculum is more commonly known as the postsubiculum and is of interest because it contains head direction cells, which are responsive to the facing direction of the head.

RECIPROCATION, RECURSION

If you haven’t yet gathered, these regions reciprocate  -meaning that in addition to feeding information forward, they cycle a continuous stream of information back and forth so that they can perceive changes in state in the stream and create the perception of constancy.

So now, let’s take direction, distance, boundaries, spaces, and head direction and look at what we need from a world model.

WORLD MODEL –

+PlACE

+BOUNDARIES AND DISTANCES

+SPACES(GRID)

+OBJECTS (SPACE OCCUPIERS)

OBJECTS IN MOTION

+DISTANCE TRAVELLED (ESTIMATED)

+SPEED,

+TIME

+DIRECTION OF TRAVEL IN ENVIRONMENT,

+DIRECTION OF TURN IN ENVIRONEMNT

+ORIENTATION IN INVIRONMENT

+LIMB MOVEMENT

+HEAD DIRECTION

+EYE DIRECTION

+FIELD OF VIEW (SPATIAL VIEW)

+SYSTEM OF MEASUREMENT (ACTION)

So we have most of our measurements.

MATCHING ORIENTATION, POSITION, SPACE AND LOCATION

Now lets look at a movie of a rat and a single neuron firing in his entorhinal cortex.

(A) PLACE CELLS. [ show movie ]

The hippocampus contains unique type of pyramidal neurons called place cells. Place cells exist throughout all hippocampal areas including the entorhinal cortex.

They form within minutes after an animal enters a novel location in an environment, firing action potentials only when an animal enters that specific location, and are maintained until the environment changes.

These place cells store and recall “maps” of novel locations.

We have mentioned several times now that we experience and learn the world as sequences of sensory signals, and this  applies to cognitive maps which are also learned sequentially, one location at a time; So, it could be reasonably extrapolated that place cells do not just store and recall cognitive maps, they store and recall novel sequences in general, whether they be visual sequences constructed by seeing novel locations, or auditory sequences constructed by hearing novel songs.

Two properties of place cells are worth mentioning.

The first is that they rely upon chemical receptors which allow rapid, synaptic modification – and that’s in contrast to the the neocortex which relies more upon voltage-dependent calcium channels which promote slow, stable synaptic modification.

As result of this fast synaptic adaptation, hippocampal place cells modification synapses only minutes after they start firing.

The second property of place cells is that they are one of the select types of neurons in the brain that may be replaced by new neurons.

With the ability to produce place cells, the hippocampus has the ability to create fresh templates required for storing additional novel sequences as they are learned over a lifetime.

So, by combining rapid synaptic modification with neurogenesis, place cells provide an ideal means of rapidly storing and recalling potentially limitless numbers of novel sequences learned by the CA3 autoassociative network.

Within minutes of learning them and for the next several hours, these novel sequences may be accessed as episodic memories (memories of personally experienced events, dependent upon the place and time of their learning);

The details of the situation where the memories were learned are retained.

Emotions, which we will cover in a later lesson, determine the probabiilty of our rehearsal of an episode, and therefore it’s saliency (importance) and information density (content).

GRID CELLS OR SPACE CELLS

[ SHOW MOVIE HERE ]

Now lets look at another video.

A grid cell is a type of neuron that allows them to understand their position in space – the Neural Basis of Spatial Navigation. Grid cells get their name from the fact that connecting the centers of their firing fields produces a triangular grid

over time, these firings accumulate, form a set of small clusters, and the clusters form the vertices of a grid of loosely equilateral triangles. This regular triangle-pattern is what distinguishes grid cells from other types of cells that show spatial firing.

By contrast, if a PLACE CELL from the hippocampus is examined in the same way, then the dots build up to form small clusters. Generally, there’s only one cluster (one “place field”) in a given environment, and even when we see multiple clusters, there is no regularity to their organization. They are truly place dependent.

So different cells for places and spaces. one map for place, one for spaces

Grid cell activity does not require visual input, since grid patterns don’t change when the lights are turned off.

But When visual cues are present, they exert strong control over the alignment of the grids: Rotating a cue card on the wall of a cylinder causes grid patterns to rotate by the same amount. In other words, landmarks cause grid patterns to rotate.

fundamentally we can thik of all matching everywhere in the isocortex, as following this general rule just as we described with polarizing lenses in geometry.

Grid patterns appear on the first entrance of an animal into a novel environment, and usually remain stable thereafter. When an animal is moved into a completely different environment, grid cells maintain their grid spacing, and the grids of neighboring cells maintain their relative offsets.

In contrast to a hippocampal place cell, a grid cell has multiple firing fields, with regular spacing, which tessellate the environment into a hexagonal pattern.

The grid cell along with Head direction cells, Border cells, speed cells and Place cells provides commensurability between our body and movement in relation to the environment

Many species of mammals can keep track of spatial location even without visual, auditory, olfactory, or tactile information, by integrating their movements—the ability to do this is referred to in the literature as path integration. A number of theoretical models have explored mechanisms by which path integration could be performed by neural networks.  The principal ingredients are (1) an internal representation of position, (2) internal representations of the speed and direction of movement, and (3) a mechanism for shifting the encoded position by the right amount when the animal moves.

Because cells in the EC encode information about position (grid cells[1]) and movement (head direction cells and conjunctive position-by-direction cells[13]), this area is currently viewed as the most promising candidate for the place in the brain where path integration occurs. However, the question remains unresolved, as in humans the entorhinal cortex does not appear to be required for path integration.

Researchers showed that a computational simulation of the grid cell system was capable of performing path integration to a high level of accuracy However, more recent theoretical work has suggested that grid cells might perform a more general denoising process not necessarily related to spatial processing.

I’m going to translate that as we have become so visually dominant that we either do not develop or have lost the ability to recall distsances without use of counting as do blind people.

Researchers suggested that a place code is computed in the entorhinal cortex and fed into the hippocampus, which make the associations between place and events that are needed for the formation of memories.

A HEXAGONAL LATTICE.

Grid cells have firing fields spread across the entire environment (in contrast to place fields which are restricted to certain specific regions of the environment)

The firing fields are organized into a hexagonal lattice

… and are, at least ideally, equally spaced apart, such that the distance from one firing field to all six adjacent firing fields is approximately the same (though when an environment is resized, the field spacing may shrink or expand differently in different directions; Barry et al. 2007)

Firing fields are equally positioned, such that the six neighboring fields are located at approximately 60 degree increments

The grid cells are associated with, and therefore anchored to, external landmarks, but the persist in darkness, so that grid cells may be part of a self-motion based map of the spatial environment.

This is what you’ll find in the literature. However, that’s becasue they were using evenly sized spaces to test the rats. The triangles deform to produce a place that is about the size of the bodily movement of the animal.

THE LATTICE IS A THREE DIMENSIONAL SPHERE

The lattice is a three dimensional sphere.

The lattice consists of triangles because …. Well, that’s a quantum thing. Cheapest Best.

THe triangles organize into hexagons because … Well, that’s a quantum thing. Cheapest Best.

The result spherical field consists of hexogons is rather spherical or wheel shaped

The size of the sphere or hexagonal space is determined by the size of the space we can move through with our bodies using the movement possible with our limbs.

There is a limit to the prediction of the distance by these spaces.

That limit is extend-able with practice.

TRINANGLES, TETRAHEDRONS, AND HEXAGONS

… are the optimum data representations for optimum choice. Every living creature on Earth uses the genetic code, in which DNA stores information using four nucleotide bases. The sequences of nucleotides encode information for constructing proteins from an alphabet of 20 amino acids.

the way DNA is assembled inside cells. In this situation, the molecular machinery inside a cell must search through the molecular soup of nucleotide bases to find the right one. If there are four choices, a classical search takes four steps on average. So the machinery would have to try four different bases during each assembly step.

But a quantum search using Grover’s algorithm is much quicker. When there are four choices, a quantum search can distinguish between four alternatives in a single step. Indeed, four is optimal number.

This thinking also explains why there are 20 amino acids. In DNA, each set of three nucleotides defines a single amino acid. So the sequence of triplets in DNA defines the sequence of amino acids in a protein.

But during protein assembly, each amino acid must be chosen from a soup of 20 different options. Grover’s algorithm explains these numbers: a three-step quantum search can find an object in a database containing up to 20 kinds of entry. Again, 20 is the optimal number.

In other words, if the search processes involved in assembling DNA and proteins is to be as efficient as possible, the number of bases should be four and the number of amino acids should to be 20—exactly as is found. The only caveat is that the searches must be quantum in nature.

This general idea is the same for triangles, hexagons: least cost search. least chance error.

Afaik the most likely theory of the sub-particle universe consists of tetrahedrons.

Now that we’ve got the basics down, Let’s go thru some text on the rest of the different types of cells.

BOUNDARY OR BORDER CELLS

Boundary cells (also known as border cells or boundary vector cells) are neurons in the hippocampal formation that respond to the presence of an environmental boundary at a particular distance and direction from us. These cells, with these firing characteristics were predicted on the basis of properties of place cells, and the Boundary cells were discovered in several regions of the hippocampal formation: the subiculum, presubiculum and entorhinal cortex.

In the medial entorhinal cortex, border/boundary cells comprise about 10% of local population, and are intermingled with grid cells and head direction cells. During development the border cells (and Head Directions cells but not grid cells) show adult-like firing fields as soon as rats are able to freely explore their environment at around 16-18 days old. This suggests HD and border cells, rather than grid cells, provide the first critical spatial input to hippocampal place cells.  Which is what we would expect.

SPATIAL VIEW FIELD OF VIEW CELLS

Spatial view cells were found in the CA1 region, the parahippocampal gyrus, and the presubiculum.

Spatial view cells are used for storing an episodic memory that helps remember where a particular object was in the environment. Spatial view cells enable them to recall locations of objects even if they are not physically present in the environment. These spatial view cells do not only recall specific locations, but they also remember distances between other landmarks around the place in order to gain a better understanding of where the places are spatially.

In real world applications, monkeys remember where they saw ripe fruit with the aid of spatial view cells. Humans use spatial view cells when they try to recall where they may have seen a person or where they left their keys. Primates’ highly developed visual and eye movement control systems enables them to explore and remember information about what’s present at places in the environment without having to physically visit those places. These sorts of memories would be useful for spatial navigation in which the primates visualize everything in a worldly manner that allows them to convey directions to others without physically going through the entire route.

Other cells stopped responding when the monkey looked toward the normally effective location in the environment if the view details were obscured. These cells were in the CA3 region of the hippocampus. The results indicate that for CA3 cells, the visual input is necessary for the normal spatial response of the neurons, and for other cells in the primate hippocampal formation, the response still depends on the monkey gazing toward that location in space when the view details are obscured.

These latter cells therefore reflect the operation of a memory system, in which the neuronal activity can be triggered by factors that probably include not only eye position command/feedback signals, but also probably vestibular and/or proprioceptive inputs. This representation of space “out there” is be an appropriate part of a primate memory system involved in memories of where in an environment an object was seen and more generally in the memory of particular events or episodes for which a spatial component normally provides part of the context.

In other words, we are processing multiple competing spatial estimation systems which consist of both the bodily (pro-prio-perceptive) memory, the visual prediction of the space, and objects in the visual field.

We are taking a lot of measurements using purely relative measures of any stable index.

HEAD DIRECTION CELLS

Head direction (HD) cells are neurons found in a number of brain regions that increase their firing rates only when the animal’s head points in a specific direction. They provide our “sense of direction”. When the animal’s head is facing in the cell’s “preferred firing direction” these neurons fire at a steady rate (i.e., they do not show adaptation), but firing decreases back to baseline rates as the animal’s head turns away from the preferred direction (usually about 45° away from this direction).[7]

HD cells are found in many brain areas, including the cortical regions of postsubiculum (also known as the dorsal presubiculum), retrosplenial cortex,[8] and entorhinal cortex,[9] and subcortical regions including the thalamus (the anterior dorsal[10] and the lateral dorsal[11] thalamic nuclei), lateral mammillary nucleus,[12] dorsal tegmental nucleus and striatum. It is thought that the cortical head direction cells process information about the environment, while the subcortical ones process information about angular head movements.[13]

A striking characteristic of HD cells is that in most brain regions they maintain the same relative preferred firing directions, even if the animal is moved to a different room, or if landmarks are moved. This has suggested that the cells interact so as to maintain a stable heading signal (see “Theoretical models”).

The system is related to the place cell system, located in the hippocampus,[15] which is mostly orientation-invariant and location-specific, whereas HD cells are mostly orientation-specific and location-invariant. However, HD cells do not require a functional hippocampus to express their head direction specificity.[16] They depend on the vestibular system,[17] and the firing is independent of the position of the animal’s body relative to its head.[18]

Some HD cells exhibit anticipatory behaviour:[19] the best match between HD activity and the animal’s actual head direction has been found to be up to 95 ms in future. That is, activity of head direction cells predicts, 95 ms in advance, what the animal’s head direction will be. This possibly reflects inputs from the motor system (“motor efference copy”) preparing the network for an impending head turn.

HD cells continue to fire in an organized manner during sleep, as if animals were awake.[20] However, instead of always pointing toward the same direction—the animals are asleep and thus immobile—the neuronal “compass needle” moves constantly. In particular, during rapid eye movement sleep, a brain state rich in dreaming activity in humans and whose electrical activity is virtually indistinguishable from the waking brain, this directional signal moves as if the animal is awake: that is, HD neurons are sequentially activated, and the individual neurons representing a common direction during wake are still active, or silent, at the same time.

SPEED CELLS

running speed is represented in the firing rate of a ubiquitous but functionally dedicated population of entorhinal neurons distinct from other cell populations of the local circuit, such as grid, head-direction and border cells. These ‘speed cells’ are characterized by a context-invariant positive, linear response to running speed, and share with grid cells a prospective bias of ?50–80 ms. Our observations point to speed cells as a key component of the dynamic representation of self-location in the medial entorhinal cortex.

Speed cells are present in the medial Entorhinal cortex. They form a part of a larger set of neurons that are involved in cognitive mapping of the surrounding environment. The other neurons are located in the entorhinal cortex or the hippocampus. They include Place cells, Grid cells, Boundary cell and Head direction cells. It is known that the speed cells respond to the speed of the animal’s movement and is not dependent on external factors such as visual surroundings.

The speed cell firing in response to the movement of the animal provides instantaneous running speed to the Grid cell. The grid cell in turn uses this information along with the head direction in order to calculate the location of the animal in the cognitive map. [1]

The speed cells fire in response to variations in speed of the animal. It was also found that unlike place cells, the speed cells are independent of visual cues. Darkness did not influence the firing rate of the animal. Another interesting feature of the cells is that the firing of the cells is better correlated with the future speed of the animal suggesting that the speed of the animal is known in advance by the speed cells.

The grid cell along with Head direction cells, Border cells, speed cells and Place cells provide a correlation between different movement aspects of the animal with respect to its environment.[2]

INDEX-MAKING

A place code of some sort is computed in the entorhinal cortex and fed into the hippocampus, which can make  associations between places and times – meaning events – that are needed for the formation of memories and subsequent iterations of higher level memories.

The researchers are working on the exact structure of the three dimensional data, and aren’t quite sure how three dimensional space is calculated other than perhaps by head direction.

So, like our art piece and lenses, think of flashbulb photographs of fragments that the hippocampal formation organizes primarily by spatial references.

THREE POINTS

We remember fragments in the brain locations where we calculate them

We organize them geometrically in to a simulation or model of space we can act within.

Memories are composed and organized as episodes.

Location is the most general index to the episode.

But once an episode is created,

Because of sparsity, storage of episodes is infinite.

Our memory is however plastic so forgetting is a natural process unless rehearsed.

Generalization alone will modify memory.

Continuous stream of information

Hierarchy of Time Loops

“no observer, no state, just recursion”

Memories are created and strengthened by maintaining activity. This is called long term potentiation.

The only memories you can depend upon are procedural (bodily motions)

SO LETS UPDATE OUR WORLD MODEL – WE PRODUCE CELLS THAT PRODUCE INFORMATION THAT TELLS US:

WORLD MODEL –

+PlACE LOCATION, LANDMARK

+BOUNDARIES AND DISTANCES

+SPACES(GRID)

+OBJECTS (SPACE OCCUPIERS)

OBJECTS IN MOTION (attention)

+DISTANCE TRAVELLED (ESTIMATED)

+SPEED,

+TIME

+DIRECTION OF TRAVEL IN ENVIRONMENT,

+DIRECTION OF TURN IN ENVIRONEMNT

+ORIENTATION IN INVIRONMENT (in relation to place)

+HEAD DIRECTION

+EYE DIRECTION

+FIELD OF VIEW

+SYSTEM OF MEASUREMENT (ACTION)

So we more than satisfy the necessary abilities of (1) an internal representation of position, (2) internal representations of the speed and direction of movement, and (3) a mechanism for shifting the encoded position by the right amount when the animal moves.

IDENTIFYING AND ENCODING NOVELTY

the sensory stream flows up the cortical hierarchy as far as it has to go until all of it has been matched with memory networks along the various levels of the neocortical hierarchy;

at the same time, predictions from the neocortex flow down the hierarchy trying to match past experiences stored in the inventory of neocortical memory networks with the current information carried in that sensory stream.

But there are many more projections carrying information down the hierarchy than up the hierarchy – which means that predictions from the neocortex form the overwhelming bulk of this “up and down” view of processing in the neocortical hierarchy.

But that doesn’t explain how the brain recognizes novel aspects of sensory information, such as a new cafe visited, a new person met, or a new book read; since these things are novel, there are no neocortical memory networks able to match them.

Such novel sensory information flows right up the neocortical hierarchy, remaining unrecognized and unmatched even by IT until it gets to the top, which perhaps surprisingly is not neocortex at all – it is the hippocampus.

So, the hippocampus lies at the top of the neocortical hierarchy, where it processes the novel sensory information from the world only after the neocortex has failed to recognize that information.

But more importantly, the hippocampus only processes unmatched sensory information after the rest of the neocortex has unsuccessfully tried to associate, match, and organize it into a model of the world;

INFORMATION FLOW

[Show Diagram]

Now lets cover how information flows through the hippocample formation.

Information flows one direction in a loop that starts in the neocortex, flows to the parahippocampal, and parrirhinal, and finally into the entorhinal cortex, then passes through a circuit consisting of the dentate gyrus to CA3 to CA1 to the subiculum.  And there is an extensive network of recursive CA3 to CA3 connections. …

… then continues to the fornix and back to test, merge average, sychronicity (which I think is processed as certainty) and then to the mammilary bodies, and then to the thalamus, then the cingulate gyrus, then the cingulusm, and then information returns to the entorhinal cortex – maintaining ‘state’ or ‘perception of continuity of experience’.

So, The hippocampus represents the end of the sensory stream processing line. But the hippocampal formation consolidates at a minimum, the sensory motor information, organizes it, and then merges with the next moment of stimulation coming from the neocortex, and repeats the cycle of integration.

That, in data processing terms, is the brain’s ‘main loop’, or ‘application loop’, where moment by moment your sense of ‘you’ experiences and forgets a constant stream of information that is disambiguated into objects and contexts, associated and matched with other objects and contexts, collected into the hippocample formation, combined into an episodic experience, and if novel, saved and if not, reinforced.

If you are ever unfortunate enough to be in a medical state where there is nothing going on other than awareness of awareness – you will understand that the ‘root you’ is a few half seconds of memory in the deepest regions of the hippocampus aware of nothing other than ‘waiting’ for something to happen that never does, but thankfully you can’t remember for more than a few half seconds.

BUT WHAT INFORMATION ARE WE TALKING ABOUT????

I think in our current age we tend to imagine the brain is sending packets of information around like telephones or computer networks. But that isn’t really true. All the information is saved wherever it is calculated. ANd instead  …. connections fight for attention and ‘food’

So it’s more like organizing complex networks of christmas lights, and whatever strings of christmas lights stay lit create our mental experience.

(connections, relations, associations, and frequency)

everything is calculated and stored where it’s calculated.

But how is this matching done?

And how are memories ‘indexed’ so that we can discover them?

Well, it’s pretty interesting really.

Our brains produce a global unique id for us – a GUID-like set of structures.

And it’s done with place cells everywhere in the brain.

Every column and every analogy to it consist of a what and where.

SUMMARY

So that completes our understanding of the first cycle of information processing, consolidation, indexing.

AND THAT”S THE FIRST IDEA, OR BIG REVEALS ORGANIZATION BY ….GEOMETRY,

(CURT TELL STORY ABOUT CARVING SPACE OUT OF WATER WITH PREDICTIONS OF OUR bODILY MOVEMENTS)

FEATURES > OBJECTS > MODELS

– locations (grid cells everywhere in the neocortex – everywhere) location relative to YOU. That’s the big novelty.

– spatial reference frames for every column, object, etc.

=========== MEMORY ===================

HIPPOCAMPAL REGION PART II – MEMORY

=========== MEMORY ===================

MEMORY

Now we’re moving on into how we store that indexed data into memories.

THE BRAIN NEEDS VOLUME TO CALCULATE

SIZES OF REGIONS ARE NOT INDICATIVE OF MEMORIES STORED OR CALCULATIONS PERFORMED THERE

MEMORIES ARE PRODUCED BY SEQUENCES (Think a line)

MEMORIES ARE STORED WHERE THEY ARE CALCULATED (as sequences)

MEMORIES ARE SAVED BY REHERSAL

MEMORIES ARE INDEXED BY ASSOCIATION

MEMORIES ARE RECALLED BY ANY ELEMENT OF THE ASSOCIATED NETWOR

LONG TERM POTETIATION

(before attention we have ltp)

WHAT WE KNOW ABOUT MEMORY SO FAR:

THREE FORMS OF MEMORY

The most interesting in the study of these various memories is that we find the great psychological structures:

1) The unconscious Self based on procedural memory, provides automatic behaviors, offers to consciousness information to evaluate.

2) The conscious Observer, with its attention organized around the short-term memory, judge the coherence between actions and results, reworks and adjusts unconscious mechanisms.

3) The biographical Self, long-term memory providing a temporal dimension to the psyche, presents to the conscious Observer the state of existence project, data assembled on the time axis whose evaluation produces anticipation.

MEMORY FORMATION

  The Hippocampus functions as a hub intgrating in a widespread network of memory, navigation and the perception of time. The EC is the main interface between the hippocampus (now) and neocortex (past and future). The EC-hippocampus system’s role is in the formation declarative memories and in particular spatial memories including memory formation, memory consolidation, and memory optimization in sleep.

During wakefulness, the hippocampus encodes novel sensory information;  First, the CA3 autoassociative network learns novel sequences, which can be accessed as working memories, and second, place cells store and recall the novel sequences, which can be accessed as episodic memories.

During non-rapid eye movement (NREM) sleep, the hippocampus consolidates novel sensory information by transferring the invariance of newly encoded novel sequences into neocortical memory networks, which can be accessed as semantic memories.

So during wakefulness you fill up a day’s worth of short term memories in your hippocampus.

  During non-rem sleep  you’re purging short term memories from your hippocampus to your neocortex.

  During Rem sleep means you’re re-adjusting your optical circuitry like a printer calibrating itself.

  And your thalamus and brain stem allow you to sleep by shutting down the majority of sense perception that comes through it, letting your neocortex do what it will.

ROLE SUMMARY

The hippocampus and neocortex are complementary memory systems, with the hippocampus being used for rapid, “on the fly,” unstructured storage of information involving activity arriving from many areas of the neocortex, while the neocortex would gradually build and adjust the semantic representation on the basis of continuously accumulating information.

The hippocampus organizes experience into mental maps, and the time, place, direction and orientation.

If we return to our piece of art for a moment, and imagine that not just the man… but whatever number of objects and spaces (Positive objects and negative spaces) that we have disambiguated from the background (the full scope of our vision and other senses), all in parallel, matched by their association, and organized by location, orientation, position, direction, speed and time.

MEMORY  – HOW ARE MEMORIES STORED????? —— HERE

SPARSITY – VERY LITTLE INFORMATION IN EACH NOVEL EXPERIENCE – LIMITLESS EXPERIENCES

(but not limitless disambigutiy)

MEMORY

Spike Memories generate network intensity.

Memory is stored in this association of fragments, as sequences (comparable) not snapshots (incomparable)

Competition between series

MEMORIES STORED WHEREVER THE DATA IS GENERATED, AND THEN ALL ASSOCIATED.

THE NEXT BIG IDEA, OR BIG REVEAL –

MOMENT BY MOMENT CIRCUIT (something to compare against – the smallest unit of comparison / competition )

Hippocampal spatial representations are encoded as sequences during behavior

10 times a second (100ms at about 30cm distance)

“where was I, where am I know, and where am I going to be…. even where I choose I am going to be.””

So we produce a three point line so to speak ten times a second.

So our spatial cognition works very much like splines on a curve if you’ve worked with splines in drawing.

Hippocampal place-cell sequences paths to remembered goals Brad E. Pfeiffer & David J. Foster… Nature, 2013

rewards…

hoppocampus likes to replay goal locations … dopamine neuroms asscociated with goal.

HIERARCHY OF TEMPORAL CIRCUITS –

back upward to the neocortex, to the organization of actions that you take, to the series of actions that you take, to short term memory to long term memory.  a hierarchy of timings.

…The unit of measure is you (action)…..   memories are goal reward biased.

sleep process is the same.  but without reward bias.

HIERARCHY OF SEQUENCES

back upward.  “There is no observer, other than memory of an observation. There is no comparison other than to previous sequences. there is no order other than that created by sequences. there is no existence perceived, without change in time.”

TOGGLING BETWEEN EXPERIENCE, PREDICTION, AND REMEMBERING

THE THALAMUS TELLS THE CORTEX HOW FAST TO EXECUTE IT’S SEQUENCES BY FREQUENCY (OCILLATION)

(which is offset by the ability of the boutons to supply resources – creating demand for more and larger boutons with more and larger surface area, with more and larger supply of neurocehmicals, to act faster and more frequently.)

THE HIERARCHY OF THE BRAIN FROM THE SENSORY TO PREFRONTAL IS JUST A HIERARCY OF INCREASINLY DENSE (COMPLEX) SEQUENCES OVER LONGER PERIODS OF TIME.

THE CEREBELLUM CALCULATES TIMING

PROCESS SUMMARY

So:

New hippocample cells can form, creating new potential unique series that function as identifiers of time and space.

And, like all other neural cells they remember sequences as sparse and unique.

those  sequences consists of the equivalent of a unique time and space key or index.

the novel unmatched inputs of the experience are associated with that key sequence.

The novel experience is associated with a place code.

The novel experience and place code is associated with a space code (grid cell map)

Non novel experiences are reinforced by retaining attention (a closed circuit)

So, by analogy, we have a global unique identifier, new data creation, update (or revise), read, and not so much ‘delete’ as ‘decay’.

And by analog our brains remember three dimensional space because we can act in three dimensional space, but with a bias to forward forward right and forward left – three faces of a hexagon. Three sides of a triangle.

I am not clear as to whether the hippocampal provisions remain or are issued remembered, and dismissed. This may be because no one knows, or because the subset of the literature I’ve read doesn’t answer the question. My suspicion is that ‘it depends’ because some locations and places are used a lot and others are not. And I’m cautious because of the cellular structure and tissue volume in these regions, which makes me think it’s ‘only as long as its getting used’. (I’ve sent out a few emails on this question)

RESULT

Action and Memory

As far as I know

– short term memory is constructed in the Hypothalamus.

– is persisted by the

– Thalamic (attention),

– Cortical (disambiguation),

– Entorhinal (integration)

– hippocampal (retention)

… cycle or loop.

– Lasts 18 to 30 seconds which decays in a sort of half life or logarithmic process – through replacement with new information.

– Is like internal vision and internal voice, is highly variable between individuals

– Allows us to task switch, not multi-task, within a window.

– Persists and is consolidated through non-rem sleep.

– May or may not be lost from short term memory after persistence by rehearsal.

– And in any active cerebral network

– reinforces by use or repetition

– encodes novelties by repetition

– converts to long term memory during non-rem sleep (which might be the reason for our need for sleep)

– (during rem sleep we are adjusting sensory timing but we will cover that in a later video.)

– All variations in memory are caused by the following dimensions:

– Attention

– Novelty

– Emotion

– Rehearsal

– Re-use

… of the neurons, columns, regions, networks, lobes co-stimulated.

– Consisting of:

Identity Provisioning

                – Rapid Neural Maturity (or genesis)

                – Unique Sequences

Position

– Location And Routes (Paths)

                – Objects

– Spaces (grid)

                – Auto Associations (memories)

Movement

– Speed,

– Distance

– Time

Orientation

– Direction of movement In Environment,

– Direction Of Turn,

– Head Direction

– Eye Direction

State

        – Recursion (18-30 seconds)

State Change

                – Various recursive information time spans.

                – (Back-propagation)

Really. And what scares me is how little brain volume it takes to do all that

COMPUTER GAMING

Now if you were writing a 3d computer game, or if you’ve ever played one, you probably can identify all those variables that are used to produce your sensations.

It’s not magic.

That’s why computer games work – they’re giving you what you already have.

In fact, one of the problems we have today, is that games are so much better than your eyes, that we have to blur certain motions so that you don’t either break the illusion, or get ill when watching.

[ DIMENSIONS ]

So let’s review our dimensions

Sensory Nervous System > Thalamus > columns > regions > networks > hemispheres > Cortex (integration)

Eye > Thalamus > Primary Visual Cortex > dividing into Dorsal(Scene) and Ventral(object) Pathways > feeding into PARAHIPPOCAMPAL(scene) Perirhinal(object) > entorhinal (place and location) > Dentate Gyrus (episode) > Body orientation > Limb orientation > > Direction of Movement > Direction of Turning > Head Direction > Eye Direction

Landmark (location) > Place > Objects (Positive) > Spaces (Negative)

Episode (Memory) = Unmatched Object and scene > Novelty (Novel Series) > Grid (Space) > Location (place, Landmarks) > Objects(volumes) >  Scenes (backgrounds)

PURPOSE

  So that you can act in the real world.

  Imagine that you can carve 3d world out of a block of space with your senses and actions. that’s a little closer to what we’re doing.

  But are we building a model of the world?  Not so much.  We are building a model of how we may act in the world. it’s not a model of the world per se. It’s a model of predictions of how our senses bodies and limbs ca interact with that world

  So and so the standard of measurement, the unit of measurement in that model of predictive actions, is your physical, sensational, emotional, and intellectual capabilities.

  We observe the world as it is.

  We produce a model of how to act in it, at our scale of capable action – which is why in childhood spaces seem much bigger.

  And we overlay that model on the world as spatial, sensory, social, emotional, and intellectual measurements.

  And that should make the world a lot more sensible to you.

  You see what the camera sees. Your brain disambiguates it into actionable objects, spaces, and boundaries, and ignores whatever isn’t actioanble, and draws your attention   to anything that might be.

  It’s wonderous really. 😉

OUR SIXTH SENSE IS OUR MOST NECESSARY: SPATIAL

1. Somewhere, we do have a sixth sense and that’s our spatial perception.

2. So that we can act in real time.

3. Where our candidate movements and candidate objects and spaces are produced at the same time.

4. Where we can let free those movements to our body (wow!)

5. Where we can let free those events, locations, objects, sensations, emotions to our attention.

6. Where our attention can make use of our frontal cortex

7. Where our frontal cortex can problem solve and plan novel and complex thoughts and actions.

CLOSING

Well, at this point  you should have an understanding of how this fabulous neural machine at the top of our nervous system does it’s magic – at least as far as our unconscious is concerned.

I hope you enjoyed this topic.  Next we’re going to decrease our resolution a bit and walk through the major circuits of the brain and see how they work in concert to produce a continuous stream of experience we call consciousness.

This is Curt Doolittle for the Propertarian Institute.

HOMEWORK

Inferotemperal cortex

Perihippocampal cortex

Perirhinal cortex

Entorhinal Cortex

Subiculum

Dentate Gyrus

Hippocampus

Fornix

Spatial View (Field of View) Cells

https://en.wikipedia.org/wiki/Spatial_view_cells

Head Direction Cells

https://en.wikipedia.org/wiki/Head_direction_cells

Boundary Cells

https://en.wikipedia.org/wiki/Boundary_cell

Place Cells

https://en.wikipedia.org/wiki/Place_cell

Place Fields

https://en.wikipedia.org/wiki/Place_cell#Place_fields

Grid Cell (Space Cell)

https://en.wikipedia.org/wiki/Grid_cell

Speed Cells (movenent of the body in space)

https://en.wikipedia.org/wiki/Speed_cells

EXTRA CREDIT

Grover’s Algorithm 😉

https://en.wikipedia.org/wiki/Grover%27s_algorithm

============================================================================

==========================

—THE CIRCUIT—

==========================

[ WIRING DIAGRAM ]

Now, sorry, but I’m going to show you a scary picture.

It’s a wiring diagram of the brain. at the top we can see the cortex, layers, and types of neurons, and at the bottom we see the:

 – Hippocampus (experience, short term memory, and encoding novelties)

 – Thalamus (attention)

 – Basal Ganglia (context and releasing to action)

 – Cerebellum (memory timing and coordinating)

 – Amygdala (emotions)

 – Metencephalon (brain stem)

================

— ATTENTION —

================

Now let’s discuss attention.

This is the thalamus.

And…

This is a view of the thalamocoritical connections, there are about ten times as many returning from the cortex as entering it.

[ RELATION BETWEEN COLUMNAR LAYERS AND THE THALAMUS]

0. At the very lowest levels, We lean to predict. We are competing with the universe between the rate of our prediction of the future and the time before it comes about. We identify and seize opportunities for gains, or avoidance of losses by acting to change the course of events for our benefit.

The first iteration is disambiguation resulting in a representation of the world.

Where what we mean when we say prediction is “disambiguation”. We sense a multitude of information all of which is ambiguous EXCEPT in relation to all other sensations within a narrow window of experiential time.

We aren’t necessarily identifying a picture so to speak, we are predicting a feature, object, model, or entity. we remember images later on in the iterative cycle of cognition, in episodic memory. Episodic memory keeps an associative network alive long enough to allow it’s reinforcement and future reconstruction.

When you learn. you need to accumulate less and less information in order to make the same prediction.

The second iteration is association and ‘valuation’.

The third iteration is prediction of possibilities

the fourth iteration is focus on possibilities

the fifth iteration is incremental pursuit or planning how to pursue those possibilities.

And we release instructions to our body at any point along that to react now, not react now, wait, act, or plan.

Do you see how the via-negativa keeps appearing everywhere? disambiguation, competition, constraint of information,

But over the course of these iterations, somehow we must keep the neural circuits active so that the hierarchy stays in place while we’re using it. Because otherwise we will be distracted by some other shiny idea, and another, and another, and another.

And that is done by the thalamus.

1. The thalamus regulates attention, attention feeds circuits, and fed and unfed circuits change accordingly

2. Fragments, features, objects, models, entities, scenes, and sequences are stored along with their POSITIONAL RELATIONS, or what we call GEOMETRY. While we’re covering a lot of material that may or may not be familiar to you, this is something we will see continually throughout our studies, where geometry serves as the constant relation across increasingly complex concepts.

3. Fragments, features, objects, models, entities, scenes, and sequences are stored in a hierarchy, with very small things at the back and bottom and larger things more complex things at the top and front.

4. We process information in that hierarchy:

– first we compose a model of the world of our attention,

– then we associate that model of the world of our attention,

– then we predict some futures models of the world realities of our attention,

– then we focus our attention to those future worlds,

– then we focus our attention on how to bring those worlds about,

– then we perform actions to bring those worlds about,

5. Meanwhile every one of those steps is continuously recycling and continuously changing, and continuously providing feedback, in response to the continuous stream of information coming from the world, our compositions, associations, futures, focuses, actions.

6. And we us our thalamus to regulate our attention

– internal or external,

– now or later,

– narrow or wide, –

and any one of those above steps”

– information collection and composition,

– associations,

– future,

– focuses,

– actions …

So that our columns in the cortex put their effort into whatever purpose we find most rewarding.

7. In the literature the thalamus ‘excites’ the cortex but this isn’t quite correct. The thalamus suppresses information, and lets other information trough, so it’s more correct to say it grants access to information every where along the hierarchy that is valuable. Then the SYMPATHETIC nervous system excites the body if necessary. That means that while we really cannot multi task so much as task switch within the tolerances of our short term  memory, the thalamus can distribute awareness – at a cost – across the brain and the body can increase resources to the brain via the blood stream if necessary.

The example I like to use is getting in the zone, juggling, plate spinning, then doing math in your head on another. trying to remember a series of numbers and letters.

8. And we keep alive those cortical circuits that were used in each iteration, so that attention is preserved, associations preserved, predictions are preserved, and neural connections are grown – and grown because they are preserved.

9. predictions from neocortical memory networks constitute the bulk of thalamocortical processing as evidenced by the fact that there are ten times as many feedback projections from the neocortex to the thalamus as there are the other way.

10. Predicting what is about to occur in the world is the main function of the neocortex; rather than simply sensing and responding to the world, the neocortical memory network inventory forms a model of it that it uses to retrieve memories acquired from past experiences so that they can be combined with current sensory information to make a new, modified prediction about a specific situation (Hawkins, 2004).

11. Lets take the example I found elswhere, of swinging a tennis racquet. As you swing it, visual information describing where the racquet is and how fast it is travelling is carried up the hierarchy within the sensory stream and fed forward to your neocortex.

At the same time and based on past experiences, your neocortex feeds back its predictions about where the racquet should be and how fast it should be travelling;

these predictions travel down the hierarchy and feed back onto the sensory stream at every level.

If the recent sensory information and the experience-based predictions “match” well enough, your brain will follow through with the swing, using previous experiences plus the specific current sensory information, to make the best swing for that specific situation.

If the recent sensory information and experience-based predictions do not match, however – say the racquet suddenly breaks mid-swing and now feels off-balance – the swing may be interrupted as the sensory stream will not match the predictions, resulting in your attention being diverted to the issue.

[ SO FOR COMPOSITION – BUT THE RESULT IS ACTION ]

So far we’ve talked about representation…

but … relationshpbetween visualization and action.

[ CHAINS AND SEQUENCES OF ACTIONS ]

[ PARALLEL ACTIONS ]

[ TIMING THOSE ACTIONS]

[ RELEASING THOSE ACTIONS]

IMAGINATION (COMPOSITION FROM FRAGMENTS AND PREDICTIONS)

Conversely, By its ability to fully retrieve and manipulate any working, episodic, or semantic memory in the brain, the hippocampus can combine these memories together any way it wants resulting in imaginary mental scenes.

That is the source of imagination and our control over imagination.

PREFRONTAL CORTEX

OUTTAKES

VARIATIONS IN EXPERIENCE

Each of us seeks a steady state we intuit as normal, or resting.

Our attention is biased to rest in

sensory, physical, social, or higher reasoning

Our experience is biased to

Felt Experience

Observed information

Associated information  (free association)

Imaginary representation (daydreaming)

Imagining others imagining (sympathy and empathy)

Planning representation (sortition, thinking)

Reasoning representation (reason)

Calculative representation (transformation)

Linguistic Representation (language)

Symbolic Linguistic Representation (language, mathematics)

Calculative Symbolic Representation (software, simulations, etc)

DUE DILIGENCE – LIMITS – PREVENTING MISUNDERSTANDING.

Now, the argument that we see a simulation can be a problem we want to avoid.

1. we see what the camera sees.

The primary visual difference between living creatures is just

– the spectrum of light

– The scale of object and scene formation

– The salience of objects and scenes

– The scale of the model(simulation)

– The depth of the locations and pathways

– The learning and forgetting curve

2. The differences between humans is in

– salience of objects or spaces

– balance between perception, association, imagination, prediction, planning.

3. We use short term hippocample memory on the one hand, and predictions from the neocortex on the other and construct a simulation of negative spaces and positive objects

4. So we ‘pull useful information’ out of the world we see, hear, and feel. It’s not that we don’t capture all the information that our eyes, ears, taste, smell, and touch do. It’s that we only convert salient information into a spatial model so that it is accessible to our attention, and that we must learn what is salient.

5. Because if we can’t act in relation to it – it doesn’t matter. Hence why we don’t see ultraviolet.

6. And we can bias our perception to others, objects and experiences (women) or spaces, objects and processes (men) – and we do.

Personal Note:

When I first started working seriously on AI in the mid 1980’s, the data structure i used was before during and after events, a time, location, space, byte-map of changes in states (actions and consequences), and a byte-map of associated emotional values.

I kept trying to harmonize that with what I knew of neural networks at the time (which really hasn’t been improved much from my perspective other than speed and scale because of technology) and those networks were just so slow and tiny at the time it wasn’t possible to simulate anything useful.

In retrospect, I was going the right direction. I did that probably because it was as close to what I was doing writing computer adventure games, and was the only solution I could think of that produced a trivial data structure. Now I understand. That’s what we do. But machines were so slow and serialized it didn’t matter. Today’s speeds were too far off in the future.

I would say that knowing that evolution and our solution to the problems of simulation are so similar, we should look at the brain for what it must be doing given our understanding of the problem of creating simulations.

==========================

—[ BASAL NUCLEI ]——-

==========================

The cerebellum is involved in the coordination of movement. A simple way to look at its purpose is that it compares what you thought you were going to do with what is actually happening down in the limbs, and corrects the movement if there is a problem. It is also involved in motor learning (eg. learning to ride a bike.)

The basal ganglia are involved in the control of complex patterns of movements. Many of their functions are actually still a mystery in neuroscience. However, it is known that they inhibit “inappropriate” movements. A lesion to these areas releases the movements that the ganglia inhibit (think Parkinson’s or Tourette syndrome).

Working memory

The basal ganglia has been proposed to gate what enters and what doesn’t enter working memory. One hypothesis proposes that the direct pathway (Go, or excitatory) allows information into the PFC, where it stays independent of the pathway, however another theory proposes that in order for information to stay in the PFC the direct pathway needs to continue reverberating. The short indirect pathway has been proposed to, in a direct push pull antagonism with the direct pathway, close the gate to the PFC. Together these mechanisms regulate working memory focus.[20]

Decision making

Two models have been proposed for the basal ganglia, one being that actions are generated by a “critic” in the ventral striatum and estimates value, and the actions are carried out by an “actor” in the dorsal striatum. Another model proposes the basal ganglia acts as a selection mechanism, where actions are generated in the cortex and are selected based on context by the basal ganglia.[29] The CBGTC loop is also involved in reward discounting, with firing increasing with an unexpected or greater than expected reward.[30] One review supported the idea that the cortex was involved in learning actions regardless of their outcome, while the basal ganglia was involved in selecting appropriate actions based on associative reward based trial and error learning.[31]

The basal nuclei is a collection of subcortical gray matter surrounding the thalamus. It has four separate parts.I won’t cover them. iT’s unnecessary for our level of resolution.

Somewhere in our evolutionary history it served as an “ancient motor system”, but that over time has been replaced by the neocortex. However, it retains a necessary role in neocortical data processing.

(2) The Nuclei’s circuits are arranged in parallel and convergent – they’re organized like a funnel.

(3) The circuits process context to alter the matching thresholds for partially matched memory networks.  So that the most suitable memory networks for a particular experience can be converted into movements or made avalable for cognition.

(4) matching reflects the degree to which neocortical memory networks correctly predict current information about the world carried within the sensory stream; matching can be good, partial, or poor.

(3) BN circuits possess two pathways, the striatonigral (STN) and striatopallidal (STP) pathways, and the relative activities of these two pathways determines the speed and accuracy of neocortical processing by establishing a universal, dynamic matching threshold that is calibrated during rapid eye movement (REM) sleep; increasing the STN pathway lowers the matching threshold whereas increasing the STP pathway raises the matching threshold.

The matching threshold is consistent across the neocortex. It is also dynamic in that it can loosen or tighten the threshold and let more impulses through or not – think just letting your body impulsively do what it wants when you’re relaxed vs concentrating on something that requires a great degree of full body, slow, precise motion.

(9) A memory network will only be released to movement or cognition if it meets a matching threshold The matching treshold is rougly the percentage of minicolumns in a memory network that must be individually matched.

When the matching threshold met, all the information contained by that memory network is excited in a process called autoassociation.

As an example of auto-association, consider that while on safari you see an elephant head behind a large bush – right away, your neocortex predicts that there will be an entire elephant, even though you do not actually see the entire elephant. From experience, your neocortex knows that elephant heads are almost always accompanied by the rest of the elephant, so despite being given partial sensory information your neocortex uses autoassociation to make a statistically likely prediction that there will be an entire elephant.

(4) When it is not clear which one out of several competing memory networks should be converted into a movement or cognition, BN circuits convergently parallel process context to modify the matching thresholds for each of the competing memory networks, allowing the most statistically suitable memory networks for a particular situation to be autoassociated and converted into movements or cognitions.

(5) Over time, BN circuits employ reinforcement learning to adjust the relative strengths of the STN and STP pathways, maintaining the matching threshold at an ideal set point.

As each additional piece of context is matched and converted into recognition, the information converges at the striatum and passes down BN circuits in parallel; in doing so, the statistical impact of each piece of context is weighted and used to modify the matching thresholds for each of the competing motor memory networks. Pieces of context that make a particular motor memory network more statistically suitable lower the matching threshold for that memory network (making it easier to match and convert into movement), whereas pieces of context that make a particular motor memory network less statistically suitable raise the matching threshold for that memory network (making it harder to match and convert into movement). At some point, the matching threshold is lowered enough for one of them – in this case, the “step” motor memory network – such that the matching threshold is surpassed and autoassociation occurs,

By incorporating context into memory network matching, BN circuits allow the statistically best memory network for the job to be converted into a movement or cognition. It is important to remember that it is the neocortex, not the basal nuclei, that contains the memory networks; the basal nuclei helps choose the best memory network for the job but it does not produce movements or cognitions, as evidenced by studies showing that the basal nuclei process information late in the initiation phase of a movement

As noted earlier the BN circuits have a convergent structure and are arranged in parallel with respect to each other. Regarding their convergent structure, the greatest degree of convergence occurs as the neocortex and other structures project onto the striatum; consequently, the striatal medium spiny neurons are able to incorporate many different pieces of information from widespread areas of neocortex allowing them to act as “coincident detectors” (Groenewegen, 2003) or more specifically, context detectors.

So the Basal Nucleus activates context, to assist in matching sets of features, objects, models, entities, to scenes, by detection of coincidence on one hand and conflict on the other.

For our purposes, lets recall that

SUMMARY

==========================

—[ TIMING ]——-

==========================

Cerebellum – Memory Timer

Striking points:

(1) The cerebellum contains contains more neurons than all of the other brain structures put together and, almost entirely through the olives, forms cerebellar circuits with them.

(2) The cerebellum consists of massively repeating groups of neurons called modules lying within these cerebellar circuits, the most prominent of which are vestibulocerebellar (VC), spinocerebellar (SC), and cerebrocerebellar (CC) circuits; CC circuits, which link the cerebellum with the neocortex, are particularly prominent.

(3) The cerebellum is needed not only for high-level movements, but high-level cognitions as well.

(4) The main processing role of CC circuits is memory timing – while neocortical memory networks retrieve memories as sequences to produce movements or cognitions, cerebellar timing networks time the sequences resulting in practiced, high-level movements and cognitions.

(5) With practice, CC circuits employ supervised learning to adjust the strength of the timing signal and modify timing networks.

==========================

—[ HIPPOCAMPUS ]——-

==========================

(1) The hippocampus is a small infolding of cerebral cortex that is strongly interconnected with the neocortex.

(2) The hippocampus lies at the top of the neocortical hierarchy, processing novel sensory information from the world only after the neocortex has failed to recognize that information.

(3) During wakefulness, the hippocampus encodes novel sensory information; first, the CA3 autoassociative network learns novel sequences, which can be accessed as working memories, and second, place cells store and recall the novel sequences, which can be accessed as episodic memories.

(4) During non-rapid eye movement (NREM) sleep, the hippocampus consolidates novel sensory information by transferring the invariance of newly encoded novel sequences into neocortical memory networks, which can be accessed as semantic memories.

(5) By its ability to fully retrieve and manipulate any working, episodic, or semantic memory in the brain, the hippocampus can combine these memories together any way it wants resulting in imaginary mental scenes.

===================

—CONSCIOUSNESS—

===================

THALMO CORTICAL RESONANCE

Recurrent thalamo-cortical resonance is an observed phenomenon of oscillatory neural activity between the thalamus and various cortical regions of the brain. It is proposed by Rodolfo Llinas and others as a theory for the integration of sensory information into the whole of perception in the brain.[1][2] Thalamocortical oscillation is proposed to be a mechanism of synchronization between different cortical regions of the brain, a process known as temporal binding.[3] This is possible through the existence of thalamocortical networks, groupings of thalamic and cortical cells that exhibit oscillatory properties.

Thalamocortical oscillation involves the synchronous firing of thalamic and cortical neurons at specific frequencies; in the thalamocortical system, the exact frequencies depend on current brain state and mental activity. Fast frequencies in the gamma range are associated with much of conscious thought and active cognition. The thalamus in this system acts as both the gate for sensory input to the cortex as well as the site for feedback from cortical pyramidal cells, implying a processing role in sensory perception in addition to its function in directing information flow. The state of the brain, whether it be conscious, in REM sleep, or non-rapid eye movement sleep, changes how sensory information is gated through the thalamus.

[ TIMER ]

To tie back to our 3d world simulation model, In 3d video games, there is a system timer or clock, and this timer is how events are synchronized. In multi-player games the problem of timing is most evident. If you’ve ever played a game ad it seems like you were shot before  you could see the other person, it’s because your predicted position on his computer, and the server clock matched.

EVOLUTIONARY THEORIES OF CONSCIOUSNESS

Theories of consciousness have been linked to thalamocortical rhythm oscillations in TC-CT pathway activity. One such theory, the dynamic core theory of conscious experience, proposes four main pillars in support of conscious awareness as a consequence of dorsal thalamic activity:

1. the results of cortical computations underlay consciousness

2. vegetative states and general anesthetics work primarily to disrupt normal thalamic functioning

3. the anatomy and physiology of the thalamus implies consciousness

4. neural synchronization accounts for the neural basis of consciousness.

This area of research is still developing, and most current theories are either partial or incomplete.

[ SEARLE’S ARGUMNET – ACTUALLY, IT”S SEMANTIC ]

As an ending note, searle’s chinese room argument seems to have been solved, meaning that the brain’s semantic content consists of networks of position, motor, sensory relations. The same would be true for an artificial intelligence, using ACTTIONS to compute rather than symbols. in other words, change the chinese characters into flip books of human actions with various numbers of pages. in other words, the vocabulary and gammar are networks of position, sensation, motor information.

Assuming a computer had the same relationship with it’s sensory(input) motor(output) systems then the characters have meaning.

So organisms an machines with the same body type and mind, would produce a similar grammar of action, and a similar grammar of speech; or those that communicate in an alien language but translate into a natural grammar of sensory-motor equivalences, would be indifferent.

[ REVIEW – DIMENSIONS ]

0. I won’t repeat the review of neurons here. I’ll (Hopefully) add the time index for both summaries in the lesson text.

1. HIERARCHY

(Dendtritic spikes ? , dendrites ?, synapses 1000-5000 trillion in adults) ->

…. Neurons(100b with 16b cortex) ->

…. …. Columns 100-200m ->

…. …. …. Layers (6-10)

…. …. …. …. Modules 2M ->

…. …. …. …. …. Regions (52/hemisphere) ->

…. …. …. …. …. …. Hemispheres (left and Right) ->

…. …. …. …. …. …. …. Cortex.

…. …. …. …. …. …. …. …Thalamocorical Projections

…. …. …. …. …. …. …. Thalamus

(REPEAT OF BASIC PROCESSING UNIT SECTION)

1. The column is the basic processing unit of the neocortex.

2. Columns disambiguate stimuli into predictions of sequences of stimuli.

3. Columns remember correct predictions of sequences, not individual sensations like a camera.

4. Columnar layers divide the labor of producing and predictions in time and space.

4. Neurons in columns compete to produce a prediction of a relations, fragments, features, then object.

4. Columns accumulate predictions to produce a prediction (meaning theory) of an object or set of candidate objects

5. Columns compete or vote to produce a prediction of an object.

6. All of this occurs very fast, in massive parallel, in iterations, accumulating, holding, and discarding information depending upon the success of the prediction, by completion of a circuit over time in a continuous recursive feedback loop.

if you have followed me long enough to pick up on my use of the “chomskyism” of Continuous Recursive Disambiguation, this is where it’s coming from. When we speak we are just imagining and planning meaning describing it, moving forward in time, and repeating until we get our point across.

4. We process information in a hierarchy:

– first we compose a model of the world of our attention,

– then we associate that model of the world of our attention and value the possibilities

– then we predict some futures models of the world realities of our attention – possibilities or demands,

– then we focus our attention to those future worlds,

– then we focus our attention on how to bring those worlds about,

– then we perform actions to bring those worlds about,

– And we release instructions to our body at any point along that to react now, not react now, wait, act, or plan. So the reason you’re fast at some things is betcause by the time youi’re aware of the possibiity of taking an action the action stream was compiled for oyu along with the prediciton – you just have to ‘release’ that stream to your body to execute it.  That’s why ‘enisoning what you’re going to do’ is so important. Envisoning is in fact acting. if you want to get good at something keep envisioning doing it correctly.

6. And we us our thalamus to regulate our attention

– internal or external,

– now or later,

– narrow or wide, –

In any of the iterative steps:

– information collection and composition,

– associations and valuation,

– future prediction,

– focuses on a solution,

– actions produce and release actions …

So that our columns in the cortex put their effort into whatever purpose we find most rewarding.

7. Whatever neural networks remain active are ‘fed’ and grow, by continuous preservation of the circuit with the thalamus

7. The cortex creates a MARKET for candidates for attention. That market always exists at all times unless we specifically shop for specific goods. and the inventory changes moment by moment generating ever new streams of opportunities assuming we’re in good meaning satisfied physical condition – or, generating an competing set of opportunities for satisfying physical demands otherwise.

8. In a coming video on the Limbic system, we’ll explain how we attribute value to those candidates for action.

[ CLOSING ]

That’s our coverage of the neocortex and it’s relation with the related organs.

Hopefully, you have developed a basic knowledge of

– how information moves through the brain

– how the organs perform functions in making it possible.

– now feature and object, space, sequence, and timing are disambiguated, organized, and synthesized as thought word and deed.

– how the process functions by competition, not identification.

– how neurons program by a single tool: if-true-then excite, inhibit, disinhibit.

– and soon  you ‘ll be able to articulate ideas using the concepts and terms we’ve addresse here.

==========

OUTTAKES

=======

METHODOLOGY

teach as obj oriented pgm lang.  class prototypes etc.

imprt global efinitions. etc

WEBSITE

Bible and Education in the history of western civlization and our ‘white man’ and theuniqueness of us as the most evolved people in history.

============

<

Was this page helpful?

Leave a Reply

. . .