Common mechanisms for learning (past), navigation(present) and dreams (future?)

Sorry for the brief(?) hiatus. I have left my day job to start a venture and so am a bit preoccupied. Hopefully, the mouse trap should benefit from the new arrangements.
Today I would like to highlight a recent study from MIT that once again highlighted the fact that the same brain mechanisms are used for envisaging the future as are used for reminiscing about the past.  The study was performed on rats and found that the rats sort of replayed their day-time navigational memories while they were dreaming. This in itself is not a new news and has been known for a long time; what they found additionally is that the rats also , sort of replayed the navigational memories/ alternatives in their head at a faster rate, to sort of think and plan ahead. This use of replaying the traces to think ahead to me is very important and cements the role of default netwrok in remebering the poast and envisaging the future.

When a rat moves through a maze, certain neurons called "place cells," which respond to the animal's physical environment, fire in patterns and sequences unique to different locations. By looking at the patterns of firing cells, researchers can tell which part of the maze the animal is running.

While the rat is awake but standing still in the maze, its neurons fire in the same pattern of activity that occurred while it was running. The mental replay of sequences of the animals' experience occurs in both forward and reverse time order.

"This may be the rat equivalent of 'thinking,'" Wilson said. "This thinking process looks very much like the reactivation of memory that we see during non-REM dream states, consisting of bursts of time-compressed memory sequences lasting a fraction of a second.

"So, thinking and dreaming may share the same memory reactivation mechanisms," he said.
"This study brings together concepts related to thought, memory and dreams that all potentially arise from a unified mechanism rooted in the hippocampus," said co-author Fabian Kloosterman, senior postdoctoral associate.

The team's results show that long experiences, which in reality could have taken tens of seconds or minutes, are replayed in only a fraction of a second. To do this, the brain links together smaller pieces to construct the memory of the long experience.

The researchers speculated that this strategy could help different areas of the brain share information - and deal with multiple memories that may share content - in a flexible and efficient way. "These results suggest that extended replay is composed of chains of shorter subsequences, which may reflect a strategy for the storage and flexible expression of memories of prolonged experience," Wilson said.

To me this seals the fate of hippocampus as not just necessary for formation of new memories, but also for novel future-oriented thoughts and imaginations.

Sphere: Related Content

Major conscious and unconscious processes in the brain: part 5: Physical substrates of A-cosnciousness

This is the fifth post in my ongoing series on major conscious and unconscious processes in the brain. For earlier parts, click here.

Today , I would like to point to  a few physical models and theories of consciousness that have been proposed that show that consciousness still resides in the brain, although the neural/ supportive processes may be more esoteric. 

I should forewarn before hand that all the theories involve advanced understanding of brains/ physics/ biochemistry etc and that I do not feel qualified enough to understand/ explain all the different theories in their entirety (or even have a surface understanding of them) ; yet , I believe that there are important underlying patterns and that applying the eight stage model to these approaches will only help us further understand and predict and search in the right directions. The style of this post is similar to the part 3 post on robot minds that delineated the different physical approaches that are used to implement intelligence/ brains in machines.

With that as a background, let us look at the major theoretical approaches to locate consciousness and define its underlying substrates. I could find six different physical hypothesis about consciousness on the Wikipedia page:

1. * Orch-OR theory
2. * Electromagnetic theories of consciousness
3. * Holonomic brain theory
4. * Quantum mind
5. * Space-time theories of consciousness
6. * Simulated Reality

Now let me briefly introduce each of the theories and where they seem to have been most successful; again I believe that though this time visually-normal people are perceiving the elephant, yet they are hooked on to its different aspects and need to bind their perspectives together to arrive at the real nature of the elephant.

1. Orch-OR theory:

The Orch OR theory combines Penrose's hypothesis with respect to the Gödel theorem with Hameroff's hypothesis with respect to microtubules. Together, Penrose and Hameroff have proposed that when condensates in the brain undergo an objective reduction of their wave function, that collapse connects to non-computational decision taking/experience embedded in the geometry of fundamental spacetime.
The theory further proposes that the microtubules both influence and are influenced by the conventional activity at the synapses between neurons. The Orch in Orch OR stands for orchestrated to give the full name of the theory Orchestrated Objective Reduction. Orchestration refers to the hypothetical process by which connective proteins, known as microtubule associated proteins (MAPs) influence or orchestrate the quantum processing of the microtubules.
Hameroff has proposed that condensates in microtubules in one neuron can link with other neurons via gap junctions[6]. In addition to the synaptic connections between brain cells, gap junctions are a different category of connections, where the gap between the cells is sufficiently small for quantum objects to cross it by means of a process known as quantum tunnelling. Hameroff proposes that this tunnelling allows a quantum object, such as the Bose-Einstein condensates mentioned above, to cross into other neurons, and thus extend across a large area of the brain as a single quantum object.
He further postulates that the action of this large-scale quantum feature is the source of the gamma (40 Hz) synchronisation observed in the brain, and sometimes viewed as a correlate of consciousness [7]. In support of the much more limited theory that gap junctions are related to the gamma oscillation, Hameroff quotes a number of studies from recent year.
From the point of view of consciousness theory, an essential feature of Penrose's objective reduction is that the choice of states when objective reduction occurs is selected neither randomly, as are choices following measurement or decoherence, nor completely algorithmically. Rather, states are proposed to be selected by a 'non-computable' influence embedded in the fundamental level of spacetime geometry at the Planck scale.
Penrose claimed that such information is Platonic, representing pure mathematical truth, aesthetic and ethical values. More than two thousand years ago, the Greek philosopher Plato had proposed such pure values and forms, but in an abstract realm. Penrose placed the Platonic realm at the Planck scale. This relates to Penrose's ideas concerning the three worlds: physical, mental, and the Platonic mathematical world. In his theory, the physical world can be seen as the external reality, the mental world as information processing in the brain and the Platonic world as the encryption, measurement, or geometry of fundamental spacetime that is claimed to support non-computational understanding.

To me it seems that Orch OR theory is more suitable for forming platonic representations of objects - that is invariant/ideal perception of an object. This I would relate to the Perceptual aspect of A-consciousness.

2. Electromagnetic theories of consciousness

The electromagnetic field theory of consciousness is a theory that says the electromagnetic field generated by the brain (measurable by ECoG) is the actual carrier of conscious experience.
The starting point for these theories is the fact that every time a neuron fires to generate an action potential and a postsynaptic potential in the next neuron down the line, it also generates a disturbance to the surrounding electromagnetic (EM) field. Information coded in neuron firing patterns is therefore reflected into the brain's EM field. Locating consciousness in the brain's EM field, rather than the neurons, has the advantage of neatly accounting for how information located in millions of neurons scattered throughout the brain can be unified into a single conscious experience (sometimes called the binding problem): the information is unified in the EM field. In this way EM field consciousness can be considered to be 'joined-up information'.
However their generation by synchronous firing is not the only important characteristic of conscious electromagnetic fields — in Pockett's original theory, spatial pattern is the defining feature of a conscious (as opposed to a non-conscious) field.
In McFadden's cemi field theory, the brain's global EM field modifies the electric charges across neural membranes and thereby influences the probability that particular neurons will fire, providing a feed-back loop that drives free will.

To me, the EM filed theories seem to be right on track regarding the fact that the EM filed itself may modify / affect the probabilities of firing of individual neurons and thus lead to free will or sense of agency by in some sense causing some neurons to fire over others. I believe we can model the agency aspect of A-consciousness and find neural substrates of the same in brain, using this approach.

3. Holonomic brain theory:

The holonomic brain theory, originated by psychologist Karl Pribram and initially developed in collaboration with physicist David Bohm, is a model for human cognition that is drastically different from conventionally accepted ideas: Pribram and Bohm posit a model of cognitive function as being guided by a matrix of neurological wave interference patterns situated temporally between holographic Gestalt perception and discrete, affective, quantum vectors derived from reward anticipation potentials.
Pribram was originally struck by the similarity of the hologram idea and Bohm's idea of the implicate order in physics, and contacted him for collaboration. In particular, the fact that information about an image point is distributed throughout the hologram, such that each piece of the hologram contains some information about the entire image, seemed suggestive to Pribram about how the brain could encode memories.
According to Pribram, the tuning of wave frequency in cells of the primary visual cortex plays a role in visual imaging, while such tuning in the auditory system has been well established for decades[citation needed]. Pribram and colleagues also assert that similar tuning occurs in the somatosensory cortex.
Pribram distinguishes between propagative nerve impulses on the one hand, and slow potentials (hyperpolarizations, steep polarizations) that are essentially static. At this temporal interface, he indicates, the wave interferences form holographic patterns.

To me, the holnomic approach seems to be the phenomenon lying between gestalt perception and quantum vectors derived from reward-anticipation potentials or in simple English between the perception and agency components of A-consciousness. this is the Memory aspect of A-consciousness. The use of hologram used to store information as a model, the use of slow waves that are tuned to carry information, the use of this model to explain memory formation (including hyperpolarization etc) all point to the fact that this approach will be most successful in explaining the autobiographical memory that is assited wuith A-cosnciousness.

4. Quantum Mind:

The quantum mind hypothesis proposes that classical mechanics cannot fully explain consciousness and suggests that quantum mechanical phenomena such as quantum entanglement and superposition may play an important part in the brain's function and could form the basis of an explanation of consciousness.
Recent papers by physicist, Gustav Bernroider, have indicated that he thinks that Bohm's implicate-explicate structure can account for the relationship between neural processes and consciousness[7]. In a paper published in 2005 Bernroider elaborated his proposals for the physical basis of this process[8]. The main thrust of his paper was the argument that quantum coherence may be sustained in ion channels for long enough to be relevant for neural processes and the channels could be entangled with surrounding lipids and proteins and with other channels in the same membrane. Ion channels regulate the electrical potential across the axon membrane and thus play a central role in the brain's information processing.
Bernroider uses this recently revealed structure to speculate about the possibility of quantum coherence in the ion channels. Bernroider and co-author Sisir Roy's calculations suggested to them that the behaviour of the ions in the K channel could only be understood at the quantum level. Taking this as their starting point, they then ask whether the structure of the ion channel can be related to logic states. Further calculations lead them to suggest that the K+ ions and the oxygen atoms of the binding pockets are two quantum-entangled sub-systems, which they then equate to a quantum computational mapping. The ions that are destined to be expelled from the channel are proposed to encode information about the state of the oxygen atoms. It is further proposed the separate ion channels could be quantum entangled with one another.

To me, the quantum entanglement (or bond between different phenomenons)and the encoding of information about the state of the system in that entanglement seems all too similar to feelings as information about the emotional/bodily state. Thus, I propose that these quantum entanglements in these ion-channels may be the substrate that give rise to access to the state of the system, thus giving rise to feelings that is the feeling component of A-consciousness i.e access to one's own emotional states.

5. Space-time theories of consciousness:

Space-time theories of consciousness have been advanced by Arthur Eddington, John Smythies and other scientists. The concept was also mentioned by Hermann Weyl who wrote that reality is a "...four-dimensional continuum which is neither 'time' nor 'space'. Only the consciousness that passes on in one portion of this world experiences the detached piece which comes to meet it and passes behind it, as history, that is, as a process that is going forward in time and takes place in space".
In 1953, CD Broad, in common with most authors in this field, proposed that there are two types of time, imaginary time measured in imaginary units (i) and real time measured on the real plane.
It can be seen that for any separation in 3D space there is a time at which the separation in 4D spacetime is zero. Similarly, if another coordinate axis is introduced called 'real time' that changes with imaginary time then historical events can also be no distance from a point. The combination of these result in the possibility of brain activity being at a point as well as being distributed in 3D space and time. This might allow the conscious individual to observe things, including whole movements, as if viewing them from a point.
Alex Green has developed an empirical theory of phenomenal consciousness that proposes that conscious experience can be described as a five-dimensional manifold. As in Broad's hypothesis, space-time can contain vectors of zero length between two points in space and time because of an imaginary time coordinate. A 3D volume of brain activity over a short period of time would have the time extended geometric form of a conscious observation in 5D. Green considers imaginary time to be incompatible with the modern physical description of the world, and proposes that the imaginary time coordinate is a property of the observer and unobserved things (things governed by quantum mechanics), whereas the real time of general relativity is a property of observed things.
These space-time theories of consciousness are highly speculative but have features that their proponents consider attractive: every individual would be unique because they are a space-time path rather than an instantaneous object (i.e., the theories are non-fungible), and also because consciousness is a material thing so direct supervenience would apply. The possibility that conscious experience occupies a short period of time (the specious present) would mean that it can include movements and short words; these would not seem to be possible in a presentist interpretation of experience.
Theories of this type are also suggested by cosmology. The Wheeler-De Witt equation describes the quantum wave function of the universe (or more correctly, the multiverse).

To me, the space-time theories of consciousness that lead to observation/consciousness from a point in the 4d/5d space-time continuum seem to mirror the identity formation function of stage 5.This I relate to evaluation /deliberation aspect of A-consciousness.

6. Simulated Reality
 

In theoretical physics, digital physics holds the basic premise that the entire history of our universe is computable in some sense. The hypothesis was pioneered in Konrad Zuse's book Rechnender Raum (translated by MIT into English as Calculating Space, 1970), which focuses on cellular automata. Juergen Schmidhuber suggested that the universe could be a Turing machine, because there is a very short program that outputs all possible programmes in an asymptotically optimal way. Other proponents include Edward Fredkin, Stephen Wolfram, and Nobel laureate Gerard 't Hooft. They hold that the apparently probabilistic nature of quantum physics is not incompatible with the notion of computability. A quantum version of digital physics has recently been proposed by Seth Lloyd. None of these suggestions has been developed into a workable physical theory.
It can be argued that the use of continua in physics constitutes a possible argument against the simulation of a physical universe. Removing the real numbers and uncountable infinities from physics would counter some of the objections noted above, and at least make computer simulation a possibility. However, digital physics must overcome these objections. For instance, cellular automata would appear to be a poor model for the non-locality of quantum mechanics.

To me the simulation argument is one model of us and the world- i.e we are living in a dream state/ simulation/ digital world where everything is synthetic/ predictable and computable. The alternative view of world as real, analog, continuous world where everything is creative / unpredictable and non-computable. One can , and should have both the models in mind - a simulated reality that is the world and a simulator that is oneself. Jagat mithya, brahma sach. World (simulation) is false, Brahma (creation) is true . Ability to see the world as both a fiction and a reality at the same time, as a fore laid stage and as a creative jazz at the same time leads to this sixth stage of consciousness the A-consciousness of an emergent conscious self that is distinct from mere body/brain. One can see oneself and others as actors acting as per their roles on the world's stage; or as agents co-creating the reality.

That should be enough for today, but I am sure my astute readers will take this a notch further and propose two more theoretical approaches to consciousness and perhaps look for their neural substrates basde on teh remianing tow stages and componenets of A-consciousness..

Sphere: Related Content

Major conscious and unconscious processes in the brain: part 4: the easy problem of A-consciousness

This is the part 4 of the multi-part series on conscious and unconscious processes in the brain.

I'll like to start with a quote from the Mundaka Upanishads:

Two birds, inseparable friends, cling to the same tree. One of them eats the sweet fruit, the other looks on without eating.


On the same tree man sits grieving, immersed, bewildered, by his own impotence. But when he sees the other lord contented and knows his glory, then his grief passes away.

Today I plan to delineate the major conscious processes in the brain, without bothering with their neural correlates or how they are related to unconscious processes that I have delineated earlier. Also I'll be restricting the discussion mostly to the easy problem of Access or A- consciousness.  leaving the hard problem of phenomenal or P-consciousness for later.

I'll first like to quote a definition of consciousness form Baars:

The contents of consciousness include the immediate perceptual world; inner speech and visual imagery; the fleeting present and its fading traces in immediate memory; bodily feelings like pleasure, pain, and excitement; surges of feeling; autobiographical events when they are remembered; clear and immediate intentions, expectations and actions; explicit beliefs about oneself and the world; and concepts that are abstract but focal. In spite of decades of behaviouristic avoidance, few would quarrel with this list today.

Next I would like to list the subsystems identified by Charles T tart that are involved in consciousness:

• EXTEROCEPTION (sensing the external world)
• INTEROCEPTION (sensing the body)
• INPUT-PROCESSING (seeing meaningful stimuli)
• EMOTIONS
• MEMORY
• SPACE/TIME SENSE
• SENSE OF IDENTITY
• EVALUATION AND DECISION -MAKING
• MOTOR OUTPUT
• SUBCONSCIOUS

With this background, let me delineate the major conscious processes/ systems that make up the A-consciousness as per me:-

1. Perceptual system: Once the spotlight of attention is available, it can be used to bring into focus the unconscious input representations that the brain is creating.  Thus a system may evolve that has access to information regarding the sensations that are being processed or in other words that perceives and is conscious of what is being sensed. To perceive is to have access to ones sensations.  In Tarts model , it is the input-processing module  that 'sees' meaningful stimuli and ignores the rest / hides them from second-order representation. This is Baars immediate perceptual world.
   
2. Agency system: The spotlight of attention can also bring into foreground the unconscious urges that propel movement. This access to information regarding how and why we move gives rise to the emergence of A-consciousness of will/ volition/agency. To will is to have access to ones action-causes. In tarts model , it is the motor output module that enables sense of voluntary movement. In Baars definition it is clear and immediate intentions, expectations and actions.
3. Memory system:  The spotlight of attention may also bring into focus past learning. This access to information regarding past unconscious learning gives rise to A-consciousness of remembering/ recognizing. To remember is to have access to past learning. The Tart subsystem for the same is Memory and Baars definition is autobiographical events when they are remembered. 
4. Feeling (emotional/ mood) system: The spotlight of attention may also highlight the emotional state of the organism. An information about one's own emotional state gives rise to the A-consciousness of feelings that have an emotional tone/ mood associated. To feel is to have access to ones emotional state. The emotions system of Tart and Baars bodily feelings like pleasure, pain, and excitement; surges of feeling relate to this.
5. Deliberation/ reasoning/thought system: The spotlight of attention may also highlight the decisional and evaluative unconscious processes that the organism indulges in. An information about which values guided decision can lead to a reasoning module that justifies the decisions and an A-consciousness of introspection. To think is to have access to ones own deliberation and evaluative process. Tarts evaluative and decision making module is for the same. Baars definition may be enhanced to include intorspection i.e access to thoughts and thinking (remember Descartes dictum of I think therefore I am. ) as part of consciousness.
6. Modeling system that can differentiate and perceive dualism: The spotlight of attention may highlight the dual properties of the world (deterministic and chaotic ). An information regarding the fact that two contradictory models of the world can both be true at the same time, leads to modeling of oneslf that is different from the world giving rise to the difference between 'this' and 'that' and giving rise to the sense of self. One models both the self and the world based on principles/ subsystems of extereocpetion and interoception and this give rise to A-consciousness of beliefs about the self and the world. To believe is to have access to one's model of something. One has access to a self/ subjectivity different from world and defined by interoceptive senses ; and a world/ reality different from self defined by exterioceptive senses. The interocpetive and exteroceptive subsystems of  Tart and Baars  explicit beliefs about oneself and the world are relevant here. This system give rise to the concept of a subjective person or self.
7. Language system that can report on subjective contents and propositions. The spotlight of awareness may  verbalize the unconscious communicative intents and propositions giving rise to access to inner speech and enabling overt language and reporting capabilities. To verbally report is to have access to the underlying narrative that one wants to communicate and that one is creating/confabulating. This narrative and story-telling capability should also in my view lead to the A-consciousness of the stream of consciousness. This would be implemented most probably by Tart's unconscious and space/time sense modules and relates to Baars the fleeting present and its fading traces in immediate memory- a sense of an ongoing stream of consciousness. To have a stream of consciousness is to have access to one's inner narrative.
   
8. Awareness system that can bring into focal awareness the different conscious process that are seen as  coherent. : the spotlight of attention can also be turned upon itself- an information about what all processes make a coherent whole and are thus being attended and amplified gives rise to a sense of self-identity that is stable across time and  unified in space. To be aware is to have access to what one is attending or focusing on or is 'conscious' of. Tarts Sense of identity subsystem and Baars concepts that are abstract but focal relate to this. Once available the spotlight of awareness opens the floodgates of phenomenal or P-consciousness or experience in the here-and-now of qualia that are invariant and experiential in  nature. That 'feeling of what it means to be' of course is the subject matter for another day and another post!

Sphere: Related Content

Major conscious and unconcoscious processes in the brain

Today I plan to touch upon the topic of consciousness (from which many bloggers shy) and more broadly try to delineate what I believe are the important different conscious and unconscious processes in the brain. I will be heavily using my evolutionary stages model for this.

To clarify myself at the very start , I do not believe in a purely reactive nature of organisms; I believe that apart from reacting to stimuli/world; they also act , on their own, and are thus agents. To elaborate, I believe that neuronal groups and circuits may fire on their own and thus lead to behavior/ action. I do not claim that this firing is under voluntary/ volitional control- it may be random- the important point to note is that there is spontaneous motion.

1. Sensory system: So to start with I propose that the first function/process the brain needs to develop is to sense its surroundings. This is to avoid predators/ harm in general. this sensory function of brain/sense organs may be unconscious and need not become conscious- as long as an animal can sense danger, even though it may not be aware of the danger, it can take appropriate action - a simple 'action' being changing its color to merge with background. 
   
2. Motor system:The second function/ process that the brain needs to develop is to have a system that enables motion/movement. This is primarily to explore its environment for food /nutrients. Preys are not going to walk in to your mouth; you have to move around and locate them. Again , this movement need not be volitional/conscious - as long as the animal moves randomly and sporadically to explore new environments, it can 'see' new things and eat a few. Again this 'seeing' may be as simple as sensing the chemical gradient in a new environmental.
3. Learning system: The third function/process that the brain needs to develop is to have a system that enables learning. It is not enough to sense the environmental here-and-now. One needs to learn the contingencies in the world and remember that both in space and time. I am inclined to believe that this is primarily pavlovaion conditioning and associative learning, though I don't rule out operant learning. Again this learning need not be conscious- one need not explicitly refer to a memory to utilize it- unconscious learning and memory of events can suffice and can drive interactions. I also believe that need for this function is primarily driven by the fact that one interacts with similar environments/con specifics/ predators/ preys and it helps to remember which environmental conditions/operant actions lead to what outcomes. This learning could be as simple as stimuli A predict stimuli B and/or that action C predicts reward D .
4. Affective/ Action tendencies system .The fourth function I propose that the brain needs to develop is a system to control its motor system/ behavior by making it more in sync with its internal state. This I propose is done by a group of neurons monitoring the activity of other neurons/visceral organs and thus becoming aware (in a non-conscious sense)of the global state of the organism and of the probability that a particular neuronal group will fire in future and by thus becoming aware of the global state of the organism , by their outputs they may be able to enable one group to fire while inhibiting other groups from firing. To clarify by way of example, some neuronal groups may be responsible for movement. Another neuronal group may be receiving inputs from these as well as say input from gut that says that no movement has happened for a time and that the organism has also not eaten for a time and thus is in a 'hungry' state. This may prompt these neurons to fire in such a way that they send excitatory outputs to the movement related neurons and thus biasing them towards firing and thus increasing the probability that a motion will take place and perhaps the organism by indulging in exploratory behavior may be able to satisfy hunger. Of course they will inhibit other neuronal groups from firing and will themselves stop firing when appropriate motion takes place/ a prey is eaten. Again nothing of this has to be conscious- the state of the organism (like hunger) can be discerned unconsciously and the action-tendencies biasing foraging behavior also activated unconsciously- as long as the organism prefers certain behaviors over others depending on its internal state , everything works perfectly. I propose that (unconscious) affective (emotional) state and systems have emerged to fulfill exactly this need of being able to differentially activate different action-tendencies suited to the needs of the organism. I also stick my neck out and claim that the activation of a particular emotion/affective system biases our sensing also. If the organism is hungry, the food tastes (is unconsciously more vivid) better and vice versa. thus affects not only are action-tendencies , but are also, to an extent, sensing-tendencies.
5. Decisional/evaluative system: the last function (for now- remember I adhere to eight stage theories- and we have just seen five brain processes in increasing hierarchy) that the brain needs to have is a system to decide / evaluate. Learning lets us predict our world as well as the consequences of our actions. Affective systems provide us some control over our behavior and over our environment- but are automatically activated by the state we are in. Something needs to make these come together such that the competition between actions triggered due to the state we are in (affective action-tendencies) and the actions that may be beneficial given the learning associated with the current stimuli/ state of the world are resolved satisfactorily. One has to balance the action and reaction ratio and the subjective versus objective interpretation/ sensation of environment. The decisional/evaluative system , I propose, does this by associating values with different external event outcomes and different internal state outcomes and by resolving the trade off between the two. This again need not be conscious- given a stimuli predicting a predator in vicinity, and the internal state of the organism as hungry, the organism may have attached more value to 'avoid being eaten' than to 'finding prey' and thus may not move, but camouflage. On the other hand , if the organisms value system is such that it prefers a hero's death on battlefield , rather than starvation, it may move (in search of food) - again this could exist in the simplest of unicellular organisms.

Of course all of these brain processes could (and in humans indeed do) have their conscious counterparts like Perception, Volition,episodic Memory, Feelings and Deliberation/thought. That is a different story for a new blog post!

And of course one can also conceive the above in pure reductionist form as a chain below:

sense-->recognize & learn-->evaluate options and decide-->emote and activate action tendencies->execute and move.

and then one can also say that movement leads to new sensation and the above is not a chain , but a part of cycle; all that is valid, but I would sincerely request my readers to consider the possibility of spontaneous and self-driven behavior as separate from reactive motor behavior. 

Sphere: Related Content

Dissociable areas of memory (in MTL): Two or three?

There has been some discussion in memory literature as to whether familiarity / novelty detection and recollection (contextual recognition of a stimulus or episodic recall) are independent processes or are the same processes, but only the memory strength varies.

In 2006, an fMRI study came around that showed that there were three dissociable areas in MTL that were associated with familiarity, novelty and recollection detection.

There have been indications that recollection, familiarity, and novelty involve different medial temporal lobe subregions, but available evidence is scarce and inconclusive. Within the medial temporal lobes (MTLs), they found a triple dissociation among the posterior half of the hippocampus, which was associated with recollection, the posterior parahippocampal gyrus, which was associated with familiarity, and anterior half of the hippocampus and rhinal regions, which were associated with novelty. Furthermore, multiple regression analyses based on individual trial activity showed that all three memory signals, i.e., recollection, familiarity, and novelty, make significant and independent contributions to recognition memory performance.

This appeared to be the established dogma to me, till I came across this new PNAS paper, which again strives to swing the pendulum back in favor of memory strengths and a single process for recollection and familiarity/novelty detection. The authors found that while a distinct group of neurons in hippocampus and anygdala was responsible for novelty and familiarity detection, recollection could just be ascertained by the strength of the neural firing of these groups of neurons. Here is the abstract of the study:

Episodic memories allow us to remember not only that we have seen an item before but also where and when we have seen it (context). Sometimes, we can confidently report that we have seen something (familiarity) but cannot recollect where or when it was seen. Thus, the two components of episodic recall, familiarity and recollection, can be behaviorally dissociated. It is not clear, however, whether these two components of memory are represented separately by distinct brain structures or different populations of neurons in a single anatomical structure. Here, we report that the spiking activity of single neurons in the human hippocampus and amygdala [the medial temporal lobe (MTL)] contain information about both components of memory. We analyzed a class of neurons that changed its firing rate to the second presentation of a previously novel stimulus. We found that the neuronal activity evoked by the presentation of a familiar stimulus (during retrieval) distinguishes stimuli that will be successfully recollected from stimuli that will not be recollected. Importantly, the ability to predict whether a stimulus is familiar is not influenced by whether the stimulus will later be recollected. We thus conclude that human MTL neurons contain information about both components of memory. These data support a continuous strength of memory model of MTL function: the stronger the neuronal response, the better the memory.

PNAS has made the article freely available, so go have a look. This is what they discuss:

We analyzed the spiking activity of neurons in the human MTL during retrieval of declarative memories. We found that the neural activity differentiated between stimuli that were only recognized as familiar and stimuli for which (in addition) the spatial location could be recollected. Further, we found that the same neural activity was also present during behavioral errors, but with reduced amplitude. This data are compatible with a continuous signal of memory strength: the stronger the neuronal response, the better the memory. Forgotten stimuli have the weakest memory strength and stimuli that are only recognized but not recollected have medium strength. The strongest memory (and thus neuronal response) is associated with stimuli that are both recognized and recollected.

One methodological flaw of the current study is that it didn't take the earlier studies showing triple dissociation into account and did not differentiate between MTL neurons based on their location within hippocampus/ amygdala. If they had distinguished based on the location, they might have found some neurons that were selectively coding for recollection. In absence of such observations I find it hard to concur that recollection is not an independent process from familiarity/ novelty detection. Recollection involves binding the familiarity/ novelty cues with other neuronal cues in MTL like neurons that code for time and place . It may be that the current study completely missed out on those integrator neurons.

Sphere: Related Content

Memory formed more easily in daytime

As per a new Nature Reviews Neuroscience research highlight , conditioning in zebrafish happened better during subjective daytime (SD) as compared to subjective nighttime (SN) and this effect was mediated by the release of Melatonin during nighttime. the authors conclude that Melatonin suppresses memory formation in Zebrafish.

Learning and memory are known to be influenced by the time of day, but the nature and mechanism of this modulation has been elusive. Now, a new study shows that melatonin, a hormone released in a circadian fashion, affects memory consolidation in zebrafish.

Melatonin release peaks during the night and falls during the day, and melatonin has been shown to affect neuronal firing in the hippocampus. The authors therefore decided to investigate whether melatonin mediates the effects of the circadian system on memory formation. They found that bathing the zebrafish in 50 M melatonin prior to SD conditioning significantly suppressed memory formation, whereas administration after conditioning or prior to testing had no effect. Furthermore, administration of a melatonin-receptor antagonist prior to SN conditioning significantly improved memory retention, as did removal of the pineal gland, the site of melatonin release.

Taken together, these results show that memory formation in zebrafish is inhibited during the night relative to the day, and that this modulation is mediated at least in part by circadian melatonin release. This might direct future research into improving mental performance in humans.

While extending the research results from zebrafish to humans may be premature, some simple studies with human subjects can confirm the effect of melatonin on human learning. and memory formation.

Sphere: Related Content

Multiple Cognitive Maps: how they are kept distinct

Readers of this blog will remember a study that had shown that there were three dissociable systems in the human hippocampal regions as relevant to declarative memory. These were the anterior hippocampus (dentate gyrus) for detecting novelty; the Posterior hippocampus (CA3 )for recollecting (or using contextual cues for recall) and the posterior hippocamal gyrus for familiarity detection. Extending these to spatial memory , one can conjecture that dentate gyrus would be involved in detecting a novel cognitive map or spatial arrangement from the older stored cognitive maps; the CA3 region will actually store these cognitive maps that provide the context using which the mice (or men ) can orient oneself; while the posterior hippocampal gyrus might be involved in detecting familiarity or whether the spatial place has been visited earlier and is familiar.

Research has indicated that indeed the CA3 region contains ' place cells ' or cells that fire when a mice is near a spatial location. Multiple such cognitive maps of the environment that the mice encounters can be stored in the hippocampus.

However, as Madam Fathom has excellently elaborated, there persisted a mystery as to how widely similar, but subtly distinct cognitive maps , were distinguished within the hippocampus. As per the above model, dentate gyrus should have a prominent role to play here detecting if a new spatial location is a novel spatial location, despite it being similar in many ways to an earlier encountered spatial location.

This is exactly what has been experimentally observed. When mice which had NMDA receptors knocked off in the dentate gyrus were put in a novel environment or context, they were unable to distinguish it from the previously learned context. Thus, these mice though capable of learning could not distinguish between contexts, as presumably their ability to detect a novel context were hampered.

To me this bodes as further evidence for the cognitive map theory and I would stick my neck and say that the mechanisms and circuits involved in spatial navigation, episodic and declarative memory are same and serve a similar function. Thus, the dentate gyrus not only detects novel words in a word list (declarative memory) , but also detects novel spatial locations (cognitive maps) and novel autobiographical events (episodic memory). Similarly the CA3 region of hippocampus codes for distinct spatial maps and distinct words an facts and also distinct autobiographical memories. Similarly posterior hippocampul gyrus may detect familiarity for both facts, episodic memories (and trouble with this may lead to Deja Vu like feelings) and spatial locations.

These multiplexed use of the same brain regions, for different types of memories, may also explain why mnemonic methods like the method of loci work excellently- as the brain regions for declarative memory are the same as for discerning one's spatial location in an environment- hence it might be computationally easy to remember lists if a associated with spatial locations or a prominent cognitive map.

Tags: cognitive map, memory

Sphere: Related Content

The faculty of Imagination: Neural substrates and mechanisms

Imagination refers to holding in mind a representation that may not be (yet) 'true' and does not necessarily reflect the facts about the external world or the Reality as of now. The act of imagination may use previous memories and a general knowledge of the world to recreate past memories or to imagine novel future events.

An article by Simon Baron Cohen, discusses the biology of imagination. Simon distinguishes between the contents of imagination , which are culturally determined; and the capacity for imagination, which is biologically grounded. He also focuses on imagination as a false or distorted representation of Reality as opposed to mere imagery, which though itself also being a mental representation, may more-or-less represent the world accurately.

Imagery may be necessary for human imagination. It has been suggested that all the products of the imagination are derived from imagery, following some transformation of the basic imagery. For example, Rutgers’ psychologist Alan Leslie, when he worked in London in the 1980s, proposed that imagination essentially involves three steps: Take what he called a ‘primary’ representation (which, as we have already established, is an image that has truth relations to the outside world). Then make a copy of this primary representation (Leslie calls this copy a ‘second-order’ representation). Finally, one can then introduce some change to this second-order representation, playing with its truth relationships to the outside world without jeopardising the important truth relationships that the original, primary representation needs to preserve. For Leslie, when you use your imagination, you leave your primary representation untouched (for important evolutionary reasons that we will come onto), but once you have a photocopy of this (as it were), you can do pretty much anything you like with it.

Thus, what Leslie contends is that the faculty of imagination involves a mechanism for making a second -order representation in the mind of a past stimulus (the imagery), in the absence of the stimulus in the here and now. Crucially, the faculty of imagination also involves the ability to modify the stimulus in such a way so that it no longer represents the original stimuli accurately. One , either creatively or mundanely , transforms some aspects of the original stimulus. to come up with something imaginary (like the concept of a unicorn). Thus, it seems there are three faculties involved- one for maintaining a second order representation of an object in absence of stimulus, another for distorting or manipulating that representation to come up with novel objects and the last for keeping this novel representation as different from the original representation to avoid confusion and loss with reality.

Leslie calls the abilities to form second-order representations and the ability to change or distort these representations as Meta-representation capacity and links this to the ability to indulge in pretend play in 9-15 months old infants and the ability for mind reading or false-belief or Theory of Mind (ToM) ability in older children (4 yr olds). The contention is that pretend play enables one to keep two copies (one primary and the second a false second-order representation) of an external object in mind simultaneously while at the same time enabling one to know that one is true and the other false or a pretense. Also pretend play involves treating one object (say a banana ) as another object (say a telephone) and thus develops the capacity to distort the second order representation of an object. This meta-representation ability in turn is the pre-requisite for imagining future outcomes and thus for successful planning in older adults.

Before I continue, let me pause and define imagination faculty in more rigorous terms for forthcoming discussions. We would be primarily concerned with the ability to shift focus from now and here to then and there and also from self to others. We would be concerned with imagination as depicted in scenarios involving people or objects with agency. Also please refer to this document by Buckner and Carrol.

Thus, the following kinds of imaginations are under preview:

1. Remembering the autobiographical past (reconstructing past memories or imagining what it felt like 'then' in the past vis-a-vis 'now').
2. Simulating the autobiographical future or Prospecting (constructing plausible future scenrie that would happen 'then' as compared to 'now')
3. Navigating (constructing a scene that is 'here' (first person perspective) to 'there' (third person perspective)) Changes in spatial perspective - seeing things/ scenes from someone else's point of view . Please keep in mind the distinction between two types of spatial point-of-view taking - one involving line tracing and the other perspective taking). It is my contention that while normal kids would rely more on perspective taking, the autistic children rely on line tracing for navigation.
4. Theory Of Mind : (constructing a representation of 'another' as opposed to self) . Thus this too involves a shifting of focus from one agent to another with the concurrent risk that the new representation may not be true.

Note that all of the above involve or are about persons or entities with agencies and their beliefs/ memories / imaginations and how they are distinct from the actual reality. This is important as the neural evidence would pertain to only this class of imaginations and would not generalize to imagining events not involving agency, for example imagining a unicorn.

In what sense might a meta-representational capacity be essential for mind-reading? Let’s define mind-reading as the ability to put yourself in someone else’s shoes, to imagine the other person’s thoughts and feelings.iv Leslie’s deeply interesting argument is that when you mind-read, you again need to quarantine your primary representations. Here’s how his argument goes. Just as your mental picture of a fish has ‘truth relations’ to a real fish in the outside world, so a belief, or a sentence, has truth relations to real events in the outside world. Thus, ‘John is having an affair with his colleague’ is a primary representation of a state of affairs, and is true if John is indeed having an affair with his colleague. But when we mind-read, we again take the primary representation (step one), copy it so that it becomes a second-order representation (step two), and can then add a prefix (step three) that completely changes its truth relations with the outside world.

Thus, we can take the primary representation ‘John is having an affair with his colleague’ (step one). We can copy it to produce an identical version ‘John is having an affair with his colleague’, except this version is tagged as being a copy or a second-order representation (step two). Finally, we can add a prefix such as ‘Mary believes that’ to the second-order representation to end up with ‘Mary believes that “John is having an affair with his colleague” ’ (step three).

Such second-order representations have unique logical properties, an insight that Leslie borrowed from the standard views in philosophy of mind. They have, to use the jargon, referential opacity. ‘I pretend that “this tea-cup is hot” ’ is true if I pretend this, irrespective of whether the tea-cup really is hot. ‘Mary believes that “John is having an affair with his colleague” ’ is true if Mary believes it, irrespective of whether John really is having an affair. According to Leslie, and I think he is right, when we mind-read (just as when we use imagination), we employ such second-order representations. I can maintain my own knowledge base (John is not having an affair) whilst representing someone else’s different (possibly false) belief (Mary believes he is).

Thus, ToM has also been placed squarely in the category of tasks requiring Imagination, as we have defined it. Simon than contends that in Autism children lack this Imagination circuit or network and have deficts with ToM in particular and Imagination in general. Thus, as per this hypothesis they must also have problems imagining future autobiographical events, problems with episodic memory and problems with perspective taking approach to spatial navigation. All this still needs to be tested.

Let me now digress a little and point to data that suggests that the module we are talking about is specific to Imagination involving Agency and not for all second-order representations. When confronted with questions that do not require an agency , the autistic people perform as well on control tasks that require false representations in the mind.

To determine whether the poor performance of autistic children is due to a specific impairment in theory of mind, Leslie constructed a control task that very closely resembles the Sally-Ann task but does not rely on theory of mind. In the control task, the children watch an actor photograph a cat that is sitting on a chair. The cat then moves to a bed, and the subject is asked to predict what the photo will show. The control task is formally similar to the Sally-Ann task in that both tasks require children to infer the contents of a representation that does not reflect the current state of affairs. The photograph task and a standard FB task were given to normal 3 and 4-year-olds and mental-age-matched autistic adolescents. The autistic children performed slightly better than the normal 4-year-olds on the photograph task but worse than normal children on the standard FB task. The normal 3-year-olds failed both the FB task and the control task. The autistic children also outperformed normal children on another control task. In the second control task, the position of a cat was marked on a map. The cat then moved, and children were asked to predict the location of the mark on the map. The results indicate that autistic kids can meet the general problem solving demands of false belief tasks, but normally developing 3-year-olds cannot.

Leslie concludes that the three year olds have a difficulty with suppressing or inhibiting their own beliefs and desires (the primary representations) and selecting the second-order representations or beliefs of Sally. Thus his hypothesis is that their performance reflects a failure of inhibition or selection processing. this is clearly tied to the third ability we have identified of keeping two representations operate. They confuse between the two, and it is my contention that this is due to their inability to hold two representations of same objects in memory at the same time. Thus, three year olds also do not have episodic memory or the capacity to hold in memory the same object representations, but with different time coordinates.
Leslie also demonstrates that this deficit in inhibition, that is age dependent, is different from the deficit in Autistic children. I suspect the deficits in Autistic people are due to their propensity to view entities with agency too as objects and thus not having any beliefs , desires or emotions.

Now let me return to the Buckner and Carrol paper. It is an important paper that shows that we use the same brain areas for Remembering the past, imagining the future, ToM tasks and Navigation. Morover, the paper shows that this default network is nothing else but the default network in brain. It is to be remembered that the default network is the network active in absence of any stimulus and one feature of Agency dependent imagination we have seen earlier is that one should be able to form representations in absence of any stimuli. Also, we have already covered research that suggest that same brain areas are used for remembering the past and imagining the future.

Let me give you the abstract of the paper:

When thinking about the future or the upcoming actions of another person, we mentally project ourselves into that alternative situation. Accumulating data suggest that envisioning the future (prospection), remembering the past, conceiving the viewpoint of others (theory of mind) and possibly some forms of navigation reflect the workings of the same core brain network. These abilities emerge at a similar age and share a common functional anatomy that includes frontal and medial temporal systems that are traditionally associated with planning, episodic memory and default (passive) cognitive states. We speculate that these abilities, most often studied as distinct, rely on a common set of processes by which past experiences are used adaptively to imagine perspectives and events beyond those that emerge from the immediate environment.

I would request that the readers go back up a little and re-read the faculties required for imagination:


They involve second-order representation without stimulus (as in the activity in default network when no tasks are being carried. This spontaneous activity in the default network suggest that this is one region that can do its work without any input data.)
Distorting, changing , modifying the representation and moving them back and forth in time and space: I believe the frontal regions are involved in combining different stimuli, sequencing and planning them to archive novel combinations.
Keeping the false/ imagined/ other-people representations distinct from primary representations: I believe medial temporal lobe is crucial here. It is implicated in learning and memory and for keeping representations of past events (episodic memory). It thus has to keep the representations distinct, so that many memories may have the same constituent objects. Thus, the medial temporal lobe, I believe is mainly responsible for keeping the representations distinct.

It is interesting to note that the default network comprises of precisely these brain areas - the PFC and other frontal areas and the medial temporal lobe along with temporal-parietal junctions implicated in ToM. Also the fronto-polar region is implicated in the shifting of perspectives from self to other, from now to then etc and may be the most affected in Autism..

Functions that shift the perspective from the immediate environment to another vantage point create an interesting challenge for the brain. We must keep track of these shifts, otherwise our perceptions would blur together. Decety and Gre´zes note that ‘reality and imagination are not confused’. A computational model of how such a process might be structured is far from being defined, but it will probably require a form of regulation by which perception of the current world is suppressed while simulation of possible alternatives are constructed, followed by a return to perception of the present. Povinelli considered this issue from a developmental perspective and noted that coordination of internal perspectives ‘paves the way for the child to sustain not simply one current representation of the self but also to organize previous, current, and future representations under the temporally extended, metaconcept of ‘‘me’’’

The Fronto-polar regions are suspected behind this ability and I suspect are the most affected in Autism.

A final set of findings suggests that the frontopolar cortex contributes to theory of mind.

Thus Autism stems from the ToM deficts in the fronto-polar regions, plus the inability to keep many simultaneous representations in mind/'memory'. I would also suspect that same region is involved in agency attribution.

What does this selective generalization mean? The combined observations suggest that the core network that supports remembering, prospection, theory of mind and related tasks is not shared by all tasks that require complex problem solving or imagination. Rather, the network seems to be specialized for, and actively engaged by, mental acts that require the projection of oneself into another time, place or perspective. Prospection and related forms of self-projection might enable mental simulations that involve the interactions of people, who have intentions and autonomous mental states, by projecting our own mental states into different vantage points, in an analogous manner to how one projects oneself into the past and future.

In the end they very wisely conclude:

In this article, we have considered the speculative possibility that a core brain network supports multiple forms of self-projection. Thinking about the future, episodic remembering, conceiving the perspective of others (theory of mind) and navigation engage this network, which suggests that they share similar reliance on internal modes of cognition and on brain systems that enable perception of alternative vantage points. Perhaps these abilities, traditionally considered as distinct, are best understood as part of a larger class of function that enables flexible forms of self-projection. By this view, self-projection relies closely on memory systems because past experience serves as the foundation on which alternative perspectives and conceived futures are built.

That brings us to my final conclusion. I believe this is further evidence for the different cultures of Schizophrenia and Autism. If Autistic have a deficit with this ToM/ Agency default network, we also know that in schizophrenics too the default network works abnormally. I presume it acts unusually in the opposite way to that in Autism- attributing more agency, involving more imagination and self-projection.

Tags: development, autism, memory, neuroscience, play

Sphere: Related Content

Simulating the future and remebering the past: Are we prediction machines?

This post is about an article by Schacter et al (pdf) regarding how the constructiveness of memories may crucially be due to the need to simulate future scenarios. But before I go to the main course, I would like to touch upon a starter: Jeff Hawkins Heirarchical Temporla Memory (HTM) hypothesis. I would recommend that you watch this excellent video.

As per Jeff Hawkins, we humans are basically prediction machines, constantly predicting the external causes and our responses to them. Traditionally, the behaviorist account has been that we are nothing but a bundle of associations- either conditioned pavlovian associations between stimuli and stimulus-response or a skinerrian association between our operant actions and environmental rewards. Thus every behavior we indulge in is guided by our memory of past associations and the impending stimulus. Jeff Hawkins refines this by postulating that we are not passive responders to environmental stimuli, but actively predict what future causes (stimuli) are expected and what our response to those stimuli may be. Thus in his HTM model, the memory of past events not only exerts influence via a bottom up process of responding to impending stimulus; but it is also used for a top-down expectation or prediction of incoming stimulus and our responses to it. Thus, we are also prediction machines constantly using our memory to predict future outcomes and our possible responses.

Now lets get back to the original Schacter article. Here is the abstract:

Episodic memory is widely conceived as a fundamentally constructive, rather than reproductive, process that is prone to various kinds of errors and illusions. With a view toward examining the functions served by a constructive episodic memory system, we consider recent neuropsychological and neuroimaging studies indicating that some types of memory distortions reflect the operation of adaptive processes. An important function of a constructive episodic memory is to allow individuals to simulate or imagine future episodes, happenings, and scenarios. Because the future is not an exact repetition of the past, simulation of future episodes requires a system that can draw on the past in a manner that flexibly extracts and re-combines elements of previous experiences. Consistent with this constructive episodic simulation hypothesis, we consider cognitive, neuropsychological, and neuroimaging evidence showing that there is considerable overlap in the psychological and neural processes involved in remembering the past and imagining the future.

As per the paper the same brain areas and mechanisms are involved in both remembering a past event and imagining a future one - and the regions involved include the hippocampus. These findings in itself are not so fascinating, but the argument that Schacter et al give for , as to why, the same regions are involved in both memory retrieval and future imaginings, and how this leads to confabulations and false recognitions is very fascinating. As per them , because we need to simulate the future events, and as the future events are never an exact replica of past events, hence we do not store the past events verbatim, but store a gist of the event, so that we can recombine the nebulous gist to create different possible future scenarios. Due to this fact (the need for simulation of future events), the memory is not perfect, and in normal individuals it is possible that they confabulate (attribute the source of their memory erroneously) or make false recognitions on memory tests like the DRM.

Fisrt a bit of background on DRM paradigm. In this test, a list of related words are presented to a subject: eg yawn, bed, night, pillow, dream, rest etc. All of these relate to the theme of sleep. Later in a recall test, when this thematically related word is presented to normal subjects, they most often say that they had encountered the word sleep earlier. However given an unrelated word like hunger, most are liable to recognize that the word was not encountered previously. What Schachter et al found was , that in those subjects that had damage to hippocampus/ other memory areas and were amnesics, this effect of confabulating the gist word was reduced. In other words, those with brain damage to memory areas were less likely to say that they had encountered the related word sleep during the original trial. this, despite their poor performance in overall remembering of old list items as compared to controls. This clearly indicates that remembering the gist vis-a-vis details is very important memory mechanism.

I believe that we should also take into account the prototype versus exemplar differences in categorization between the males and females into account here. I would be very interested to know whether the data collected showed the expected differences between males and females and hopefully the results are not confounded due to not taking this gender difference into account.

Anyway , returning to the experimental methodology, another sticking point seems to be the extending of results obtained with semantic memory (like that for word lists) to episodic memory.

Keeping that aside, the gist and false recognition data results clearly indicate that the constructive nature of memory is an adaptation (it is present in normal subjects) and is disrupted in amnesics/ people with dementia.

Thus, now that it is established that memory is reconstructive and that this reconstruction is adaptive, the question arises why it is reconstructive and not reproductive. To this Schacter answers that it is because the same brain mechanism used for reconstructing memory from gist are also used for imagining or simulating future scenario. They present ample neuropsychological, neuroimaging and cognitive evidence on this and I find that totally convincing.

The foregoing research not only provides insights into the constructive nature of episodic memory, but also provides some clues regarding the functional basis of constructive memory processes. Although memory errors such as false recognition may at first seem highly dysfunctional, especially given the havoc that memory distortions can wreak in real-world contexts (Loftus 1993; Schacter 2001), we have seen that they sometimes reflect the ability of a normally functioning memory system to store and retrieve general similarity or gist information, and that false recognition errors often recruit some of the same processes that support accurate memory decisions. Indeed, several researchers have argued that the memory errors involving forgetting or distortion serve an adaptive role.

However, future events are rarely, if ever, exact replicas of past events. Thus, a memory system that simply stored rote records of what happened in the past would not be well-suited to simulating future events, which will likely share some similarities with past events while differing in other respects. We think that a system built along the lines of the constructive principles that we and other have attributed to episodic memory is better suited to the job of simulating future happenings. Such a system can draw on elements of the past and retain the general sense or gist of what has happened. Critically, it can flexibly extract, recombine, and reassemble these elements in a way that allows us to simulate, imagine, or ‘pre-experience’ (Atance & O’Neill 2001) events that have never occurred previously in the exact form in which we imagine them. We will refer to this idea as the constructive episodic simulation hypothesis: the constructive nature of episodic memory is attributable, at least in part, to the role of the episodic system in allowing us to mentally simulate our personal futures.

I'll finally like to end with the conclusions the author drew:

In a thoughtful review that elucidates the relation between, and neural basis of, remembering the past and thinking about the future, Buckner and Carroll (2007) point out that neural regions that show common activation for past and future tasks closely resemble those that are activated during “theory of mind” tasks, where individuals simulate the mental states of other people (e.g., Saxe & Kanwisher 2003). Buckner and Carroll note that such findings suggest that the commonly activated regions may be specialized for, and engaged by, mental acts that require the projection of oneself in another time, place, or perspective”, resembling what Tulving (1985) referred to as autonoetic consciousness.

This Seems to be a very promising direction. The 'another time and place' can normally be simulated withing hippocampus that also specializes in cognitive maps. We may use the cognitive maps to not only remember past events, but also simulate new events. In this respect the importance of dreams may be paramount. Dreams (and asleep) may be the mechanism whose primary purpose is not memory consolidation; rather I suspect that the primary function of dreams is to work on the gist of the memory from the previous day, simulate possible future scenarios, and then keep in store those memories that would help and are likely to be encountered in future. Thus, while dreaming we are basically predicting future scenarios and sorting information as per their future relevance. Not a particularly path-breaking hypothesis, but I'm not aware of any thinking is this direction. Do let me know of any other similar hypothesis regarding the function of dream as predictors and not merely as consolidators.

Tags: memory, sleep, cognition

Sphere: Related Content

Body Posture affecting memory recall

First Mixing Memory wrote about it; and now Dave at Cognitive Daily was enchanted with this study that shows that if one assumes a body posture during memory retrieval, which is the same as the body posture at the time of memory formation, then the recall is better. I do like this study, and I think it is important, but am hardly surprised or overwhelmed by the results.

To explain the study in a nutshell (you are encouraged to read about the study at Mixing Memory or Cognitive Daily), the authors found that juts like some smells, sights or sounds can trigger associated memories, so too can the body posture. Now, to m this doesn't come a s a surprise because I have always been fascinated by the three other senses that are normally ignored by those who claim we have juts five senses: sight, sound, touch (includes all somato-sensory senses like pain , temperature etc), smell and taste. The three other senses that are normally ignored are Vestibular sense (or the sense of balance), kinesthetic sense (or the sense of self movement) and the proprioception sense (the sense of body position and posture).

Evidently if memory encoding uses some sort of sensory inputs to encode a particular memory, it is clear that memories would be assorted with all of the sense modalities and a trigger in any of them that is strong enough, can trigger that memory recall.

One can test this for the other sense too - the vestibular and kinesthetic - and one would find that one can recall memories better if the same vestibular or kinesthetic conditions are invoked. Experimentally one can have people dance, put them in merry-go-rounds, put them atop an elephant, let them drive, let them go up and down in the lift and ask for congruent or incongruent memory recall. I wont be surprised if the same effects are observed with the kinesthetic sense too. Maybe one of you can make this your thesis and tell me the results, so I can blog about it later!

Tags: memory, senses

Sphere: Related Content

The Mouse is dreaming that it is in a Trap!!

New research has established that mice dream and during their sleep there is a two-way dialog between the hippocampal recent day memory area and the neo-cortex that is believed to be involved in long-term memory.


The content of the mice dream is also no longer secret. In the sleep they are replaying the sequence of steps that they had executed in a maze, but in a reverse order, and in lesser time and in general are rehearsing the structure of the maze (the mouse trap). Learning, it is to be remembered, arises from these replays of fast rewinds and sleep it seems is necessary for learning.

Some quotes from the article:

During nondreaming sleep, the neurons of both the hippocampus and the neocortex replayed memories — in repeated simultaneous bursts of electrical activity — of a task the rat learned the previous day.

Earlier this year Dr. Wilson reported that after running a maze, rats would replay their route during idle moments, as if to consolidate the memory, although the replay, surprisingly, was in reverse order of travel. These fast rewinds lasted a small fraction of the actual time spent on the journey.
In the findings reported today, the M.I.T. researchers say they detected the same replays occurring in the neocortex as well as in the hippocampus as the rats slept.

The rewinds appeared as components of repeated cycles of neural activity, each of which lasted just under a second. Because the cycles in the hippocampus and neocortex were synchronized, they seemed to be part of a dialogue between the two regions.

Because the fast rewinds in the neocortex tended to occur fractionally sooner than their counterparts in the hippocampus, the dialogue is probably being initiated by the neocortex, and reflects a querying of the hippocampus’s raw memory data, Dr. Wilson said.

“The neocortex is essentially asking the hippocampus to replay events that contain a certain image, place or sound,” he said. “The neocortex is trying to make sense of what is going on in the hippocampus and to build models of the world, to understand how and why things happen.”

PS: My blog post has deliberately used words like 'dream', 'mouse' and 'traps' instead of the correct 'sleep', 'rats' and 'mazes': just to come up with a juicy headline!!

Tags: , memory, learning, sleep

Sphere: Related Content

Neurogeneisis, learning and small-world networks

Continuing this blog's recent focus on categorization, one possibility of how new items are classified has been hypothesized as either assimilitaion (adding the item to an existing schema in the feature space) or accomodation (addition of a new schema around the item in the feature space). We'll leave aside the newly introduced concept of Restructuring for this particular discussion.

Schemata, it is to be remembered, are conceptualized as nothing but a named cluster in the feature space. If we become a bit more audacious, we can posit that the clustering in the feature space is mimicked by the actual clustering/ connectivity of neurons in the Hippocampus (or the appropriate semantic memory brain module), with each neuron representing a particular item- say a neuron being a Halley Barry neuron. These neurons would not be randomly distributed- they form a small-world model with local clustering and bistability. whenever a group of neurons get activated together (and also belong to a cluster or clique), we can say that the memory of that category is activated.

Further suppose that learning and memory are crucially dependent on Neurogeneisis and new learning (of concepts ) happens by insertion of a new node (neuron in the small-world network of brain) and connecting it appropriately with other neurons.

As an example consider that all face recognition cells cluster together in the brain and the concept of face is activated by simultaneous activation of all cells of this cluster. The fact that a new visual stimulus (a novel human face of a stranger) is a face is determined by calculating the stimulus features and their difference from the prototypical/ exemplar face neurons and their features. A match so determined not only enables us to say that this new stimulus is a face (as this input would activate the face clique) , but would also give us an idea of where to place a new neuron that may encode for this new face and how to connect this with other neurons and with which other neurons.

Now whenever we encounter a novel stimulus we have two possibilities. If it matches some existing cluster / category, we encode this new memory by placing a new neuron coding for this in the region of that category in the feature space and (crucially) following preferential attachment attach it in a manner such that the probability of its linking to any other neighboring neuron is in proportion of the links that old neuron already has. (This can be readily implemented in brains as axonal connections will whither if not much functional activity happens at the synapse formed between the new neuron and the older one) . This is akin to assimilation of a new memory/ learning neuron. this method of insertion still keeps the neural net a small-world network.

Now consider the second case when the novel stimuli matches no older categories but necessitates that we form a new category if we have to represent that new item in the feature space. We need accommodation here. On the neural level this is still accomplished by inserting a new neuron, but this time the new node is not peripheral- the new neuron is a hub (category) neuron. So we use the method of copy to insert the new element. We copy the links (partially) of a neighboring hub (cluster center/ category label neuron) and use that link structure to link the newly introduced neuron in the small-world network. the network still remains scale-free and we have introduced a hub or a new category in this case.

All this seems very exciting. Some snippets from wikipedia article on scale -free networks are very relevant.

The mostly widely known generative model for a subset of scale-free networks is Barabási and Albert's (1999) rich get richer generative model in which each new Web page creates links to existent Web pages with a probability distribution which is not uniform, but proportional to the current in-degree of Web pages.

A different generative model is the copy model studied by Kumar et al. (2000), in which new nodes choose an existent node at random and copy a fraction of the links of the existent node. This also generates a power law.

Recently, Manev and Manev (Med. Hypotheses, 2005) proposed that small world networks may be operative in adult brain neurogenesis. Adult neurogenesis has been observed in mammalian brains, including those of humans, but a question remains: how do new neurons become functional in the adult brain? It is proposed that the random addition of only a few new neurons functions as a maintenance system for the brain's "small-world" networks. Randomly added to an orderly network, new links enhance signal propagation speed and synchronizability. Newly generated neurons are ideally suited to become such links: they are immature, form more new connections compared to mature ones, and their number but not their precise location may be maintained by continuous proliferation and dying off.

I am excited, what about you?

Tags: categorization, neuroscience, learning, memory

Sphere: Related Content

Categorization, Memory, small-world networks and neural architecture

In the last post I had wondered about the clustering based solution to categorization and how they may also inform us about how memory (semantic variety) is stored in brain, as semantic memory is best modeled by an associational or confectionist network.

Thus, a semantic memory based on clustering models may consist of associations between clusters or categories of information. For example one cluster may correspond to the names of countries and another to name of cities. A particular type of connection or association between these two clusters may map a relation of ----IS A CAPITAL OF ---- type where for example the fact that Paris is the capital of France is stored. For this knowledge to exist, one has to have prior notions of France is a Country and Paris is a City and on top of that an associational relation between the individual entities France and Paris belonging to particular clusters.

Much of this would be more apparent once relational models of categorization are also covered. For now let us assume that (semantic) memory itself may consist of clusters of neurons that are also interconnected. Interestingly one such neural architecture, that has also been able to simulate short-term memory has been the small-world network model. In this a large number of nodes (neurons ) are connected by edges (synapses) as in a typical random graph. These small-world networks are special in the sense that they have high clustering coefficients and low mean path length. Translated in English, this means they exhibit more than chance clustering (to enhance local processing) as well as display a small value of smallest mean path length (reflecting ease of global processing).


It is intriguing thta in the short term memory model using small-world networks simulation, the researchers found that the model could exhibit bistability, which may be crucial for memory formation. In bistability, the cluster or functional region corresponding to a particular memory can be in two states, depending on an input variable. Thus, a pulse (direction of attention) can activate/ deactivate a memory.

Crucially, it can be hypothesized that as the small-world network model of memory/ categorization is good for local-global processing as well as reflective of the actual brain and AI simulation architectures, the entire brain is a small-world network adequately categorizing and representing the sensory, motor and cognitive information and processing them.


A recent MEG based study has established the fact that the small-world network topology exists in functional sphere in the brain at all oscillatory levels (crucial for binding) and that seems very promising.

Tags: categorization, memory, AI

Sphere: Related Content

Categoristation: how to bookmark the interesting pages on the web!

In an earlier post, I had touched upon the different categorization theories that are in prevalence. One of these that was discussed in details was the prototype Vs exemplar method that was based on clustering and involved different representational methods of the categories thus derived.

This post is about how a new item is allocated to a pre-existing category. Simplistically, and in the last post this was the position I had taken, it seems apparent that by calculating the distance of a new item in feature space from the central tendencies of the neighboring clusters (the prototypes/ exemplars) one can find a best fit with one of the clusters and allocate the new item to that category.

This is simplistic as it explains fitting of new items to existing categories, but does not include any mechanisms for formation of new categories.

The analogical approach I take here is of how do I decide in which folder to add a new bookmark of an interesting page found on the web. Most probably the names I have chose for my bookmarks folders are reflective of the central tendencies (common prominent features) of all pages bookmarked in that folder. I would normally look at the new page, and also at my existing folders and see if there is a best fit. If so I juts file the new bookmark under the best-fit existing folder. Slightly extending the concept of categorization to resemble that of a schema, this is the classical case of assimilation in a schema.

However, in case the new web-page cannot be filed under any existing bookmark folder, I would usually create a new folder (with an adequate descriptive name based on the location of the web page in the feature space) and file the new bookmark under that new folder. This is akin to trying to fit in a novel item into existing clusters in the feature space, only to discover, it doesnt fit well with any cluster, but is an outlier. The best way to accommodate such an outlier , in my opinion, is to create a new cluster around the outlier. Extending this to schema, it is not hard to see that this is the classical case of accommodation and formation of a new schemata to incorporate a novel item that cannot be assimilated in existing schema.

Piaget, of course , stopped here (and so do I, sometimes, when managing my bookmarks!). but I would like to venture firth and discuss the other process that I engage in , very infrequently, to keep my bookmarks in good shape. This is what I would call reorganization or restructuring. when I restructure my bookmarks, I change the names, I move bookmarks form one folder to another , I merge bookmarks and also at times create more than a few sub folders. Also, interestingly, I delete some of the old bookmarks; while am captivated by some of the bookmarks and even forget to complete the restructuring part.

I believe that we too indulge in restructuring of our Schema/ categories periodically (it may be as frequent as daily during REM sleep) and that a crucial form of learning is not juts Assimilation and Accommodation, but also Restructuring. Also it is my contention, that we consciously remember anything only because we have actively restructured that information and embedded it in a contextual narrative. In the absence of restructuring, there can be information that can be used, but no conscious knowledge.

I plan to tie this up with the 3 factor model of memory that is emerging. One factor of the memory system uses familiarity detection (assimilation), the other novelty detection(accommodation), while the other involves conscious and contextual recollection(restructuring).

I also propose that these three factors are behind the three kinds of memory (content-wise and not duration wise). The first type of memory is semantic (or noetic)- facts like France's capital is Paris; the second is procedural (or anoetic) - learning how to drive- and is unconscious; while the third is episodic or autonoetic) - personally remembered events and feelings) . Of course memories would also differer along the time dimension- working memory, long-term memory etc. , but that discussion is for another day.

Also a brief not to myself - how this may be linked with Hughling-Jackson's theory of 3 states of consciousness and how they are differentially affected in dissociation- the autonoetic memory would be affected first- the noetic second and the anoetic or unconscious memory last in dissociation.

Returning back to categorization, this approach of adding new items either by assimilation, accommodation or restructuring is more guided my Mind-Is-A-Container metaphor. Other metaphors of mind- assuming it theory like - may yield to new and interesting views of how we form a theory-like theory of categorization. The other minor variation to above mind is a container metaphor may be using labels for bookmarks (instead of folders)- this is what Google bookmarks and del.icio are using. I haven't experimented with that approach to bookmarking extensively, so am not sure what new insights can be gained form them. For those readers, who use labels to organize bookmarks, their insights as comments, would be greatly appreciated.

Tags: learning, memory, categorization

Sphere: Related Content