Imagination & Prediction

“It is poor sort of memory that only works backwards”, said the Queen.

(Lewis Carroll, “Through the Looking-Glass”)

In Tutorial 7 we learned that the same process which directs the growth of brain hardware is later re-employed to direct concept formation (perception), the learning cycle, and the behavior of mind software. In this tutorial we have seen how the process is employed in memory.

Here is a reminder of the process in learning & memory:

Learning cycle: Concentration - observation - modeling - practice - variation -coordination

Memory processes: Receiving - perceiving - encoding - consolidation - retrieval - reconsolidation

Don't make the mistake of thinking that COMP is this process; COMP is only the employment of the process for the purposes of learning. What we want to look at here is how prediction is involved in the process.

Obviously, initial perception is an essential foundation for both learning and memory processing. Any event attended to in the here and now leads to the activation of other associated memories as a part of that perception.

This seems obvious if we think about it, because it is why we can remember things outside of the context of which they were established, and through stimuli other than those that were initially involved in the learning (such as a familiar smell) reminding us of a place and time and associated events.

What may not be so obvious is that most of this association remains unconscious. The hippo practises associational processing constantly and all events, every object or episode, is associated with related points in the inner model. That’s why, when we start understanding the inner model's (based on biology's) natural patterns of association, we can use them as a short cut to remembering anything; they form a universal mnemonic.

What is even less obvious is that perception in the here and now relies heavily on both the past (memory) AND the future (prediction). We call up concepts we know from the past and we imagine which ones the current percepts might match up to. We predict which ones are most likely to be correct and imagine them (render & project an inner image via the mirror neuron system) in order to compare the two in greater detail.

This act of 'imagining which ones are most probably correct', is prediction. We do it constantly. At every millisecond during reading these words, we are predicting what words might come nexus six! lingerie! ballroom dancing! See what I mean? When it doesn't make sense, we get confused. What we do from there depends on anxiety levels; if we are anxious we get worried, and if we are not anxious we tend to laugh at our own confusion (and humor such as Monty Python plays on this.)

This type of prediction is not just 'guessing'; or random decision like that made based on the rolling of dice. It is probability calculation, based on both inductive and deductive data. The 'dice' are loaded with our own personal associations and the full weight of knowledge and experience. Every association that we have learned about any person, place or set of circumstances is brought to the mental courtroom for current judgment of 'harmful' or 'beneficial' and the unconscious mind associated these with 'wrong' and 'right' (and this is in fact the root of congruous morality; something we shall discuss in later tutorials.)

Mammals use prediction even in order to move. The activity of head direction cells predicts, 95 ms in advance, what the animal's head direction will be.[62]

Predicting all the time increases efficiency. To avoid having to carefully take in and remember every last bit of visual or audible stimulus it encounters, the mind quickly acquaints itself with the world's predictability and redundancy, takes advantage of the fact that reality is predictable, and pays less attention to parts it can predict. [77]

We are applying unconscious prediction to all input even in order to perceive it. It is an all-pervading process and when we realize this, it is hard to think of intelligence in the old, behavior-based way. Intelligence, now that we can examine it, is not conforming to our old predictions as a behavior-based system; it is evermore revealing the properties of a predictive system, and viewing it as such enables predictions that prove true.

Imagination is either working out what's most likely to be happening (perception), what probably did happen (memory), what's likely to happen next (prediction), and and what would be the best response for directing what happens next to get things to go the way intelligence wants them to go.

The same process is working in multiple contexts: in the physiological context, cells have receivers for input signals. When there are enough points of similarity between signal and receptor, the cell responds with the associated output behavior.

In the context of human awareness (the everyday reality we are accustomed to living and moving in), human beings have senses for input signals. When there are enough points of similarity between signals and senses, the human being responds with the associated output behavior.

The mind is ultimately predicting our best imagined recommended behavior in the here & now for its own success and thriving; increasing its own ability to interact with more and more different events and situations. That's what it was designed to do, and anytime things get in the way of it doing that, things will go wrong.

This is how we are able to adapt to change and interact successfully with the unexpected; the mind makes predictions, interacts, then checks to see how things turned out. This constant feedback cycle keeps us 'tuned in' to the results of our interactions so that we can constantly assess and upgrade ourselves.

The same predictive process goes on for all data regardless of their memory or importance weighting.[64] In a sense, imagination is predicting all the time, because to give a clear picture of the here-and-now experience it has to guess what is going on out there using information only from the senses (which has formed the entire content of memory) and memory itself.

Imagination has to guess what's happening, what did happen, what might happen and what might be the best response to all of it. In this sense, imagination IS prediction. All it does is assess probablilities and present them in formats that all systems can use for processing.

So now you can see why imagination relies on a database of congruous associations for accurate prediction about reality, and why input control is so important. If imagination is predicting and remembering using data from a database full of associations that are nonsense, the results will often be nonsense, our decisions and behavior will make no sense and results will be disappointing.

So if we think that our mind is making bad decisions, it's a good idea to look at where it's getting its information from. Garbage in...

Input control and congruous association in hacking the genome

Here we can clearly see the impact of input control. Think about it. This means that whatever imagination believes is going on, that data is being used to affect changes on our genome according to what is predicted "necessary".

Getting the imagination to predict that weight loss is neccessary, for example, is a straightforward matter of altering blood chemistry so it carries the chemical signals saying "stop storing fat". This hack is possible via nutrition alone, so it's an easy one.

Not so easy would be convincing the genome that you need a tail, for example. How would you send the signals to the genome that a tail was necessary? Centuries of practice at trying to hang onto tree branches with the crack of your bum? Probably faster would be to stimulate the neuromuscular junctions around the vestigial tail (the bottom of your coccyx), but we're not recommending you try the experiment for the very reason that you may just get an unexpected result. Humans do carry the gene for a tail and now and again one has to be removed from a baby, so this is just an example okay?

Often the key to hacking genes on or off lies with frequency of input. Making a memory requires gene transcription which is turned on by your interest in something and its importance. When we imagine something is interesting and important, the brain will assume the necessity of putting it into memory for us, signal the genome, and off we go.

Timed learning explores the optimal frequency of input, and will be particular to you within a certain range. We'll look more closely into this later.

Wild thing…you make my heart sing…

The full development of wiring for the early nets of association and memory in the brain needs two kinds of experience -both full natural sensory input and motion. Motion brings with it the necessity for awareness of location, in both time and space. We need to remember where things are, as well as what they are and when they are likely to happen. Even the simplest of creatures can remember the direction of food, safety, warmth and danger –but only if it has been allowed to explore its environment in the first place. Apparent necessity is always the first cause of growth. The brain needs to think that something is needed in order to start developing it (our perceptual templates, once triggered, convince the brain of that need.)

The knock-on effects of wrong input were first noticed (as are many things in neuroscience) in the behavior of lab rats, and were one of the keys to discovering why lab rats have quite different neurophysiology, psychology, and even genetic expression, to wild rats.

The importance of spatial exploration showed up in the fact that lab rats raised in cages, despite having lots of previous practice at running mazes, still couldn’t learn a new maze faster than a wild rat (with no such previous experience). After eliminating possible hunger as motivation, this is still the case. Because you’ve studied spatial memory now, you’ve probably guessed that the wild rats have developed bigger hippos and better spatial navigation due to growing up in an environment with the correct input, and you’d be right –excellence in spatial learning & memory skills needs freedom of movement and exploration, and obviously the wild rats had gotten plenty of it.

What really made us sit up and take notice though was the discovery that wild rats totally lacked the lab rats’ famous penchant for drug addiction [notably cocaine, morphine and alcohol] and also had much lower blood/cortisol levels. Lab rats, given the choice, will self-administer neuroactive chemicals, preferring water with intoxicants to that without. Wild rats are happy to sample both, will try the drugs now and again on an irregular basis, but on the whole choose plain water 95% of the time.

This preference for being straight wanes if they are kept in cages for more than a few days, at which point their immunity also plummets and their blood cortisol rises. This makes sense; their freedom has been unaccustomedly restricted and they have become anxious in a new unknown environment. But this does not explain why the lab rats, which had never lived in the wild, had both high cortisol and the tendency to addiction right from the start.

By human standards they had a “better quality of life” than the wild rats; they were well fed, cleanly housed, disease-free and they had no predators. All their physical needs were taken care of for their whole lives, and they had exercise facilities, puzzles to solve and things to keep them occupied. Yet they were [comparatively] slow learners and they got addicted to drugs.

The lab rats were unconsciously anxious, not because their freedom had been taken away but because biology’s triggers for their natural neural development had been replaced by false ones, just as ours have. And they’d then wisely ingest anything that reduced the anxiety and helped bring on the relaxation response, much like we do.

When biology's ideas of what's 'important' are at odds with society's, as is most often the case, the conscious mind ends up believing something that contradicts with what the unconscious mind is aware of. We get incongruous association unless we stop taking one of them seriously and surpress any data that contradicts the other. If the mind is running society's ideal self data, our environment is impoverished, biology must be ignored and repressed, and this results in dysfunction, illness and decline.

The unconscious mind can detect when incongruous association is going on, but it cannot speak or express its knowledge in words; the only way it can communicate with the conscious mind is via imagination and emotion. Its "error message" for incongruous data therefore comes (in a healthy person) as alarm, suspicion that something is wrong, and concern for our own wellbeing. This should prompt us to pay closer attention to what we are doing and saying and thinking about, and intelligence soon works out what it was that didn't make sense (usually new or unfamiliar information or activities).

In an unhealthy (ie, anxious) person the result of incongruous error signaling is either anger or depression, either with or without paranoia, AND an exacerbation of anxiety as the problem fails to go away.

Basically if imagination fails to match up ANY belief, behavior, or belief system (percept) to what biology unconsciously knows to be reality (concept on inner model), and the input keeps saying that it is true, unresolved unconscious ideological dilemmas cause us to become dysfunctional and our behavior becomes dysfunctional. This is the road to decline and illness.

Biology expects the conscious mind to be intelligent (why should it believe otherwise?) It expects we will have no problem understanding the conscious scientific details of these unconscious sensorimotor basics. When we throw stuff at it like, '2+2 can equal 5', or 'magic goblins cure arthritis', the unconscious assumes we wish to make use of the process known as 'placebo effect'. If input keeps insisting that 2+2 CAN equal 5, or that these goblins are real material goblins doing real supernatural magic; and repressing any thoughts to the contrary, the unconscious has effectively had it's access to the mental internet blocked. Everything it knows is real is being denied, but there is no new evidence for why and no coherent explanation of this alternate math-bending, goblin-filled reality is being produced. In plain language, it's confused, and whenever we're confused we are vulnerable to deceit and coercion and the 'superstition glitch' (see tutorial 7).

Everything new that is considered must either be recognisable enough to fit in with concepts the mind already knows, or provide experiential proof for its validity as a new concept that fits in with everything else and makes sense according to the known. The unconscious is open minded enough to believe that previously unknown things or circumstances could indeed exist because life on earth is full of strange examples of unexpected new discoveries, but it needs enough data to fit new possibilities in with what it already knows. In other words, you have to demonstrate the maths or show it the goblin, or it will continue to fail to make sense out of the new input.

If the unconscious is blocked from presenting conflicting input and an incongruous, anxiety-based model is adopted, everything goes downhill from there. Things will seem to make sense but on analysis will not, and a sense of inability to cope and vague distress permeates every behavior.

Which model of reality the unconscious eventually adops depends totally on input. If the balance of input is from reality, the unconscious will adopt biology's congruous default model designed for our wellbeing.

If the balance of input is false, untrue or unhealthy, the unconscious will build an anxiety-based, incongruous model including the false 'fact' that it seems inadequate to cope alone, understand things or make good decisions, therefore you must rely on others to make decisions for you; for examples: teachers, parents, relatives, lovers, friends, anyone in a white coat or a suit, priests, anyone with lots of money, online agony aunts or favorite celebrities.

With an incongruous model of reality thinking for ourselves tends to be unconsciously discouraged as it tends to make us anxious when we try to 'make sense' out of life and relationships and events. According to our model they don't make sense, there is no explanation for many outcomes and no prediction of many harmful events. A lot of stuff seems to happen by accident. Quite naturally this makes us feel vulnerable, just like we would if we had physical accidents all the time.

A congruous model gives us the confidence of knowing what makes sense and what doesn't, and that's a great advantage when trying to compute real life outcomes with a prediction engine.

Rabbits in space –processing in N3

Between them from sensory input N1 & 2 process all the basics of experience. We often lump them together when discussing networks because they share so many tasks, but each has its own main area of processing, and for once this is going to be relevant, because what each network processes for perception and stores in memory is related to its core association.

N1 processes material relevant to sensorimotor memory; it specializes in collectiing and categorizing information about ourselves and our immediate surroundings; for example what physical material or object we are dealing with and what condition it is in –basically concrete material things.

Take for example our white rabbit. For her, 'concrete material things' means herself, her body, her sensations, her shelter, nest or burrow and the physical material experiences that occur there, such as sleeping, giving birth, keeping herself and her home and any baby rabbits clean and orderly, grooming herself, and avoiding any dangerous material substances.

White rabbit's network 2 incorporates the space in which she operates and events within that space (her context), her territory and environment, and behaviors of herself and others in that context, for example, looking for other rabbits to have sex with and searching for food and looking for other rabbits to have sex with and exploring new territory and watching out for predators and looking for other rabbits to have sex with (rabbits like sex).

So N2 needs spatial memory, so that she doesn’t get lost and can remember where food, dangers, or hot rabbits are likely to be.

This gives her all the processing she needs for the basics of simple animal behaviors “serene & clean” and “seek & squeak” as far as ‘stimulus-response’ goes –the movement of objects (in this case rabbits) in space is what behavior IS, even for the simplest of creatures. But it is N3 that enables us (and rabbits) to really start thinking about and making sense of the world.

By the time something even gets to be a memory, it is already two steps away from reality (it has been copied and compressed twice) and resolution has been lost. This doesn’t usually matter, because we don’t need all the intricate details of every event, so we compress to just save the relevant ones. Attention is like a Dolby system that cuts out trivia and retains what’s relevant, reducing the overall bit density while retaining the essentials.

DO IT NOW

You can test the compression:

Which pair of underpants or socks were you wearing two weekends ago? What is the exact wording of the first sentence of this tutorial?

Did you have any underwear on at all two weekends ago? What main subject is this tutorial about? -See what I mean?

Bit density in memory can be adjusted by weighting information. You can also test the difference between compression with 'light' or 'heavy' weighting:

Read this news report just once:

1. Ms Marsha Breeze-Sutherland, formerly of the Buddhist Ashram at Coolanka, and Mr Jeremy Hollander-Fische, professional musician familiar as the guitarist from rock band "Hook", today (the twenty second of March, a sunday), celebrated the birth of their first offspring; a daughter, in the village of Hardly Cross.

Now cover the page or turn away and try to write down the news report word for word, missing out any words you are not sure of.

See how much information your memory retained? It's not easy is it? This report is pretty boring if you don't know the people involved, and your memory is not particularly interested in it. Many of us even put 'Mrs' as the first word when we tried this!

Now read the news report below, just once, and repeat the procedure:

2. Ex nun & rock star Marsha & Jeremy celebrate daughter! They were hardly cross in Hardly Cross today when Jez Fische (guitarist with "Hook") and hot ex-buddhist nun Marsha welcomed their daughter with a massive free-for-all party. March 22 was chosen "Not because it's a Sunday but because it's the spring equinox", explained Jeremy.

Chances are this time you'll do much better. There is repetition (you have heard the story before) and this version has more emotional weighting. You're more likely to remember most of the facts from this one with greater ease. Even though this latter version contains more words, memort was more able to cram more information into fewer 'bits'.

So when you have anything to learn that seems boring or 'dry', add your own weighting. Write your own notes to emphasize important points and associate them with things that matter to you in real life, as though you were writing to explain to someone else why this material matters.

You can turn compression off with various hacks, among them TMS, induced- or self-hypnosis, and even just determined regular practice; and often access much greater detail. This is a technique that is part of the training for espionage, and practiced well it leads to a photographic memory. In advanced NH the eventual aim is to be able to turn it up and down, on and off as required.

For baking your noodle later on:

The mind is the ultimate prediction-engine, and it relies more than anything else on imagination. It's going to be difficult for most people to change their concept of imagination's status-shift from 'capacity for fantasy' to 'basis of all major mental functions', but it's looking like this is very likely to be the case.

A societal paranoia about creativity and imagination is merely the result of front loaders being scared of what they don't understand, just as a societal paranoia about science is rear loaders being scared of what they don't understand (ie, intellect, evidence and analysis).

Imaginative, creative people are used to being told their ideas are ludicrous or impossible, battling to do all the research without funding because its fun anyway, then watching them play out while some dude flies a plane, cures a disease or lands on the moon or something at which point everyone says it was a good idea and funds the building and selling of products it has resulted in instead of further research.

What spooks society out is that creative people just go right on being creative regardless of how much money or qualifications they haven't got (and sometimes, perversely, in inverse proportion to it). If you're a materially wealthy person with a poor imagination and no creativity at all, it can seem not only threatening but downright spooky watching someone with no resources pull ideas out of nowhere like rabbits from a hat. It's like MAGIC (and that equals woo woo, so its nonsense)...yet here it is, happening before their very eyes. And of course, their unconscious knows full well that it should be able to do it too, so it takes some real repression to avoid the ideological dilemma and the fearful question 'so why can't I?'

Consequently the number of people who were initially ridiculed for now-famous work has been legion. This problem is known about and has even been tested for [68]. So don't expect the new imagination paradigm to go mainstream anytime soon.

Let us rather examine it ourselves: Consider this:

Imagination uses images from current perception and prediction to enable memory, images from memory and prediction to enable current perception, and images from current perception and memory to enable prediction.

This interrelationship of core processes in N3 supports all cognition, and defines N3 as a major part of the brain's CPU, modulating multiple subprocesses in all networks.

We cannot perceive reality without imagining it. Virtually all unconscious thought is graphic imagery, and all conscious thought is based on its conclusions about and portrayal of reality.

All we have to do is stop "drawing" the wrong conclusions. Imagine that!