How does the brain work?

Copyright 2016 Graham Berrisford. One of about 300 papers at Last updated 26/08/2019 13:35



The brain’s ability. 1

Creating, using and sharing mental models. 1

Current speculations on how memory/thinking works. 1

How good is our memory?. 1


How clever are we?. 1

Natural intelligence. 1

Artificial intelligence. 1


“How your memory works”. 1

Memory Encoding Stage. 1

Memory Storage Stage. 1

Memory Retrieval Stage. 1

Brain state and brain entropy. 1

More research. 1

Other research. 1




The brain’s ability

Limited kinds of memory, intelligence and self-awareness are evident in primitive life forms

All animals “know” enough about themselves not to eat their own bodies.

Primitive animals are intelligent enough to sense the state of their environment, and respond accordingly.


Even insects, with tiny brains, communicate to perform complex tasks.

Honey bees can not only recognise a new instance of the type “pollen source”; they can communicate a description of it to other honey bees.


More intelligent actors can, for example:

-        rapidly and accurately perceive objects and events of many kinds

-        abstract and remember new types/descriptions from observations of objects and events

-        remember what actions work best in response to an input

-        perform those actions when they recognise a new instance of a type

-        converse with other actors using a language.

Creating, using and sharing mental models

How do we monitor and influence entities and events in our environment?

Conant’s “good regulator” theorem says we must have a model of those entities and events, however sketchy that model may be.


Regulatory system

Mental models

<create and use>    <abstract concepts from>

Humans   <monitor and influence>  Entities and events


“It is clear that the brain uses stable configurations of neuronal connection to model and process information and to initiate adaptive behaviors in response to information received.”

This quote is extended in the next section, but the biochemistry is incidental.


A mental model is an idea (be it simple or complex) encoded in the mind about a reality.

We hold mental models of things we perceive now, perceived yesterday, and envisage tomorrow.

A mental model is a partial and flawed description of a reality.

It is an imperfect truth, fuzzy edged, and only accurate enough to help us survive and thrive.


This paper goes on to indicate that nobody knows how mental models are stored and maintained.

And it seems nobody is sure if memory and thought are the same or different, inseparable or separable.

Nevertheless, we can legitimately speak of mental models without needing to explain how the brain works.

Because animals clearly do create mental models as side effects of physical sensations and of thinking.


The first evidence for mental models is that animals successfully recognise things they have met before.

And then act according to what they remember of their past experience.

The second evidence is that animals successfully share descriptions of realities.

E.g. honey bees tell other bees where the pollen is, and those bees then find it.


As social animals, we humans share mental models by sending and receiving communications.

Gesturing animatedly to an oncoming train is enough to share the mental model of that as a threat to survival.

As humans, we translate mental models into and out of verbal forms.

Spoken words translate mental models into transient sound waves – heard by a currently present audience.

Written words transcribe mental models into persistent graphical symbols – readable by future and remote audiences.

Current speculations on how memory/thinking works

It is presumed that memory/thinking evolved to help animals use past entities and events as a guide to future actions.

The mix of proposals and research later in this paper indicate it is likely that:

·       thinking involves continuous electrical impulses and processing of bio-chemicals

·       thinking involves a lot of parallel processing

·       only a fraction of the parallel processing is evident in the stream of consciousness

·       it is difficult to separate memory and thinking (see below)

·       a memory may change as a side effect of thinking

·       there is no meaning in the biological codes of bio-chemicals and electrical impulses per se.

·       a memory becomes meaningful when it is used (perhaps after thinking brings it into the spotlight of conscious thought).


Keeping a perfect and complete record of every past entity and event is not necessarily useful or efficient.

So, specific memories may be given up or converted into general knowledge (converted from episodic to semantic memories).

It may well be that we generalize from experience as part of an ongoing recall/re-consolidation process.

And that the fuzziness of mental models is the very quality that enables abstraction from particulars to universals.


So remember that our mental models are only sketchy representations of what is out there.

How bad is our memory? Probably worse than you think; see the references in the later section.

How clever are we? Probably not as clever as you think; see the references in the later section.


More happily for this work:

We can remember a huge amount of experience and information accurately enough to use it effectively.

Remembering events (as opposed to entities) is a human skill that dogs don’t have.

A deeper, more abstract way of approaching problems makes experts more successful. (Chi et al. 1981)

How good is our memory?

Probably worse than you think; memories are untrustworthy and unreliable.

Here is a brief distillation from this article

·       Transience: memories fade; unless we’re regularly bringing things to mind, they simply fade into the mist.

·       Absent-mindedness: we often fail to notice or remember something in the first place.

·       Misattribution: memories are altered when recalled: when a memory is repeatedly recalled it grows stronger but also changes, gets confused or blended with other memories.


Here are some other sources

·       How Quickly We Forget: The Transience of Memory

·       Absent-Mindedness: A Blessing in Disguise?

·       On the Tip-of-the-Tongue: Blocked Memories

·       How Memories are Distorted or Invented: Misattribution

·       When Suggestibility is a Liability: Wrongful Convictions

·       How the Consistency Bias Warps Our Personal and Political Memories

·       The Persistence of Memory

·       Our short-term memory lasts only 15-30 seconds (Peterson and Peterson, 1959)

·       Our long-term memory is unreliable (Implanting False Memories: Lost in the Mall & Paul Ingram)

·       We miss a lot, because our “The Attentional Spotlight” is highly selective (Cherry, 1953. The Cocktail Party Effect).


A key to intelligence is the ability to detect a family resemblance between things.

That is, to spot a repeated structure or behavior, to abstract a pattern or type from instances of it

Then, having done that, to recognise new instances and respond to them appropriately.

How clever are we?

Probably not as clever as you think.

·       Most of us find formal logic extremely difficult to cope with, fail even simple tests (Wason, 1968)

·       What we see tends to override what we hear, even if what we hear is right (McGurk and MacDonald, 1976).

·       We don’t recognise what we are incompetent at (Dunning Kruger effect).

·       We make irrational decisions (Cognitive bias).

·       We are strongly influenced by the wording of a proposition put to us (Kahneman and Tversky (1981).

Natural intelligence

Natural intelligence is a product of biological evolution.

To presume all animals or all humans have the same mental abilities defies everything we know about evolution.

What you do with your brain depends on the efficiency and effectiveness of your brain's bio-electro chemistry.

There is no reason to think that differences in physical appearance, musical ability or social skills are any greater or less than differences in brain power


At the age of three, Mozart could discriminate between chords and dischords.

He often spent much time at the clavier, picking out thirds, which he was ever striking, and his pleasure showed that it sounded good....”

In the fourth year of his age his father, for a game as it were, began to teach him a few minuets and pieces at the clavier....

He could play it faultlessly and with the greatest delicacy, and keeping exactly in time....

At the age of five, he was already composing little pieces, which he played to his father who wrote them down.” Wikipedia 2017


Not everybody is equipped by nature to be a Mozart.

Many adults discriminate little between chords and dischords, and about 4% of the population are tone deaf


When psychologists measure abilities in a human population; they find them spread across a spectrum with a roughly normal distribution.

There is usually a bulge at the lower end of the scale, representing people disabled by genetics or accident.


This applies to intelligence as it applies to musical ability.

"Scientists have investigated this question for more than a century, and the answer is clear:

the differences between people on intelligence tests are substantially the result of genetic differences."


The question is: How large is that "substantial" portion?

The results of twin, adoption and DNA studies suggest about 50%, however...

"Another particularly interesting recent finding is that the genetic influence on measured intelligence appears to increase over time, from about 20 percent in infancy to 40 percent in childhood to 60 percent in adulthood."


The implication is that children who start out relatively equal at primary school level gradually diverge as they grow older.

Their genetic potential constrains or enables their development

Still, what an adult develops through experience, training and dedication can be 40% of the whole, which leaves plenty of room to promote the importance of nurturing it.

Artificial intelligence

Imagine a computer program that examines many objects, and measures each in terms of three variables: height, breadth and length.

The program can look for a common pattern in those measures, and create a “type” to represent it.

It may find many things measure high on height but low on breadth and length. Let us call that a “tower”.

It may find many things measure high on height and length but low on breadth. Let us call that a “wall”.


But what to do with those types once they have been found?

The leap from there to human intelligence or any kind self-organising system is vast.

I can’t get my head around it; and can't envisage it the leap being made the next 100 years.


It is far from clear that AI will ever work like natural intelligence

However, it may increasingly mimic human intelligence.


You can read about neural networks and back propagation algorithms here

Machine learning algorithms can “learn” how to perform tasks by processing “big data” and finding patterns in it.

It’s not always easy to get all the big data; IBM buys companies with large datasets in areas like healthcare and weather.

Their “Watson” has natural-language capabilities to understand the jargon of a field and process the data better.

IBM then works with experts in different fields (oncologist, chefs, accountants) who “teach” Watson how they do their jobs.

(Of course, other companies may beat IBM in the artificial intelligence game.)


Through the process of encoding, storage, and retrieval, the human brain is able to access a lifetime of knowledge, experiences, and emotions.

We are able to clearly remember things like our child’s first steps, the smell of the ocean, or the way our first home looked because of the process of memory.

Being able to access your story and experiences is an invaluable mystery of the brain that is essential to what makes us “human”.


Your memory takes all your experiences and shapes them into the story of you.

How does it work?

“How your memory works”

This section is distilled from


“Memory is an astonishingly powerful thing.

In some senses, it is a mysterious thing.

[However] we can analyze the cycle of remembering.


Your brain is not a single filing cabinet.

Learning new information requires the brain to access information from different parts of the brain’s neural web to construct “memories.”

… a series of processes in the brain work together to provide a single output.


While this process is complex, it breaks down into three simple stages encoding, storage, and retrieval.”

Memory Encoding Stage

Encoding is the most basic component of memory.

This part of the process uses our base senses to process information.

Any visual, auditory, or olfactory stimulation filters to the hippocampus for processing.

Then, the sensory input combines into a singular experience.

This is why certain songs or smells take you back to a specific time or place.

Once the brain has combined the incoming information, it is then responsible for analyzing the information.

It must decide what information is important enough to store [long term].


Do the next two sections belong under encoding?


The neural network

The study of the brain and its mysteries is constantly changing.

However, scientists have concluded that the brain’s neural network involves 3 things: synapses, neurotransmitters, and dendrites.

Surely also the nerve cells themselves?


The (100 trillion) synapses are locations where nerve cells connect.

When an electric pulse stimulates a synapse, neurotransmitters releases “chemical messages” to different parts of the brain.
Dendrites are the areas of the brain that receive the message sent by the neurotransmitters.


This process is happening in every brain cell.

This process of connection across the brain’s network is constantly happening-- and just as quickly changing.

This “plasticity “is what allows people to grow and learn.

Every time a synapse sends information to a dendrite, their connection gets stronger.

The new stimulation then adds connections in your brain cells.

When how are new connections made?


Memory analysis
As this process runs in the background, your frontal cortex is performing an analysis.

This allows the brain to not only organize thought and memories but also adapt its structure as new experiences are added.


Thankfully, your brain is able to determine which things are worth encoding and which should be ignored.

E.g. If your brain were consciously choosing to process all sensory data you see or experience while driving, you would end up on the side of the road.


Memory Storage Stage

Once information is encoded by the frontal cortex, the brain then evaluates how and where to store the input.

This sorting process allows humans to function in daily life.

There are three types of storage in the brain: sensory, short-term, and long-term.

Does that mean encoding happens three times?


Sensory (less than one second)

Sensory memory is the shortest storage unit for outside input.

Scientists explain sensory memory as the initial impression left after experiencing external stimuli that functions as a “buffer” to filter input from our five senses.

Sensory memories last for less than a second.

They are only retained when the brain consciously acknowledges the input as important and sends the information to short-term memory storage.


Short Term Memory (twenty to thirty seconds)

This type of memory can only retain a small amount of information - around 5-7 things

The most common example is evidenced by remembering a phone number.

Short term memory effectively allows your brain to not only remember input but also access the information simultaneously.

If a short term memory isn’t moved to long term memory, it too will disappear rather quickly, which is why we forget names so often.


Long Term Memory

Your long-term memory houses the information the brain labels as “important”

This memory process is the reason we are able to grow and learn through life.

Repetition of information helps reinforce long-term memories.

As we learn new information or have different experiences, the brain sends this input to a part of your memory that is related.


Memory Retrieval Stage

Remembering something requires that you mindfully decide which memory you will bring to your conscious attention.

??? Mindfully? Don’t I also recall things without determining to do so?


The process of encoding and analyzing helps the brain recall the memory you are trying to retrieve.

??? How does encoding help recall? Surely that would be decoding?


The process of “forgetting” occurs when one part of the memory system isn’t working correctly.

??? Isn’t forgetting a natural process that slowly diminishes unused memories?


When I remember something, how does that affect the memory?

What about false memory syndrome?

Brain state and brain entropy

“Brain entropy and human intelligence: A resting-state fMRI study” Glenn N. Saxe , Daniel Calderone, Leah J. Morales Published: February 12, 2018


“It is clear that the brain uses stable configurations of neuronal connection to model and process information and to initiate adaptive behaviors in response to information received.

We use the working definition of brain state, and its role in brain functioning, proposed by Zagha and McCormick:

“We consider [brain] state to be a recurring set of neural conditions that is stable for a behaviorally significant period of time …”,

fluctuations in multiple states and sub-states result in “…a highly dynamic and complex control of network responsiveness and processing in relation to behavior”.

Zagha and McCormick conclude: “A major task for neuroscience is determining exactly how many sub-states exist and how they organize, interact and influence behavior”.


Entropy usually means disorder or randomness.

The tendency to increasing entropy is the gradual decline into disorder across the universe as a whole.

By contrast, here, entropy is the number of ways a system can be configured - the number of states to the system.

And brain entropy is the number of ways a brain state (or sub state) can be configured, which this study reports as related to intelligence.


I suspect the relationship is indirect.

Brain entropy seems to provide a measure of capacity, or potential.

Intelligence surely requires also an ability to create and use the state available.

And that surely involves fuzzy matching of two substates (created by perception or cognition) to each other, and consequential reconfiguration of them?

More research

This section is composed of snippets quoted and edited from this resource

You can check out at the wider resource catalogue (recommended by Adrian Hall) at


Encoding memories in engrams

Encoding beginning with perception through the senses

It converts a perceived item of interest into a construct that can be stored within the brain, and then recalled later from short-term or long-term memory.

An engram is a memory trace, a hypothetical biophysical or biochemical change in the neurons of the brain; no-one has ever actually seen, or even proved the existence of an engram.

The process of laying down a memory begins with attention (regulated by the thalamus and the frontal lobe) to external stimuli, and creates an engram in response.

A memorable event causes neurons to fire more frequently, making the experience more intense and increasing the likelihood that the event is encoded as a memory.

Although the exact mechanism is not completely understood, encoding occurs on different levels.

The first step being the formation of short-term memory from the ultra-short term sensory memory, followed by the conversion to a long-term memory by a process of memory consolidation.


Decoding memories - recall and thinking

During recall, the brain "replays" a pattern of neural activity that was originally generated in response to a particular event, echoing the brain's perception of the real event.

In fact, there is no real solid distinction between the act of remembering and the act of thinking.

These replays are not quite identical to the original, though - otherwise we would not know the difference between the genuine experience and the memory.

[They] are mixed with an awareness of the current situation.

Memories are not frozen in time, and new information and suggestions may become incorporated into old memories over time.

Thus, remembering can be thought of as an act of creative re-imagination.


Memories are not stored in our brains like books on library shelves, or even as a collection of self-contained recordings or pictures or video clips.

[They] may be better thought of as a kind of collage or a jigsaw puzzle, involving different elements stored in disparate parts of the brain linked together by associations and neural networks.

Memory retrieval therefore requires re-visiting the nerve pathways the brain formed when encoding the memory.

The strength of those pathways determines how quickly the memory can be recalled.

Recall effectively returns a memory from long-term storage to short-term or working memory, where it can be accessed, in a kind of mirror image of the encoding process.

It is then re-stored back in long-term memory, thus re-consolidating and strengthening it.

Other research

Much (perhaps most?) is still unknown about how the brain works; theories abound.


2014 Wikipedia

Long-term memory, unlike short-term memory, is dependent upon the construction of new proteins.


2009 Scientific American

Current thinking holds that new memories are encoded in the hippocampus and then eventually transferred to the frontal lobes for long-term storage. .

In other words, the location of a recollection in the brain varies based on how old that recollection is.


2014 Paul King:  Computational Neuroscientist, Redwood Center for Theoretical Neuroscience

At the most basic level, memories are stored as microscopic chemical changes at the connection points between neurons in the brain.

The strengthening and weakening of the synapses is how the brain stores information.

From here, the story becomes much more complex.

The precise way that long-term memories are structured and represented across billions of synapses is the subject of intense ongoing research and remains one of the great mysteries of neuroscience.


2013 Gerard Marx, MX Biotech Ltd., Chaim Gilon, Institute of Chemistry, Hebrew University, Jerusalem, Israel, ACS Chem. Neurosci., 2013, 4 (6), pp 983–993 February 18, 2013

“We propose a tripartite mechanism to describe the processing of cognitive information (cog-info), comprising the


           surrounding neural extracellular matrix (nECM), and

           numerous “trace” metals distributed therein.

The neuron is encased in a polyanionic nECM lattice doped with metals (>10), wherein it processes (computes) and stores cog-info.

Each [nECM:metal] complex is the molecular correlate of a cognitive unit of information (cuinfo), similar to a computer “bit”.



The average human brain has about 100 billion neurons and many more neuroglia (or glial cells).

Each neuron may be connected to up to 10,000 other neurons, passing signals to each other via as many as 1,000 trillion synaptic connections, equivalent by some estimates to a computer with a 1 trillion bit per second processor.

It used to be thought that the role of glial cells was limited to the physical support, nutrition and repair of the neurons of the central nervous system.

Recent research suggests that glial cells actually perform a much more active role, although the extent and mechanics of this role is still uncertain

Estimates of the human brain’s memory capacity vary wildly from 1 to 1,000 terabytes.

(cf. the 19 million volumes in the US Library of Congress represents about 10 terabytes of data).

The processes of memory encoding and retrieval [are] achieved using a combination of chemicals and electricity.

Subtle variations in the mechanisms of neurotransmission allow the brain to [perform] encoding, consolidation, storage and retrieval of memories.

The hippocampus is responsible for analyzing inputs and ultimately deciding if they will be committed to long-term memory.

It acts as a kind of sorting centre where the new sensations are compared and associated with previously recorded ones.

The various threads of information are then stored in various different parts of the brain, although the exact way in which these pieces are identified and recalled later remains largely unknown.


We usually discuss the brain as it appears in human kind.

We think of it as the place in the body where memory, intelligence and self-awareness are located.

Our mind may be primarily located in the brain, but it is also a part of the body.


For Karl Popper (1902–1994) there are three aspects of the mind–body problem: matter (material reality) minds (of actors) and creations of the mind (including descriptions).

The body–mind problem is the question of whether and how our thought processes [in our mind] are bound up with brain events [in our body matter]...

I would argue that the first and oldest of these attempted solutions is the only one that deserves to be taken seriously [namely: mind and matter interact].

— Karl Popper, Notes of a realist on the body–mind problem.


For John Searle (b. 1932) the mind–body problem is a false dichotomy; that is, mind is a perfectly ordinary aspect of the brain.

There is no more a mind–body problem than there is a macro–micro economics problem; they are different levels of description of the same set of phenomena.

But Searle is careful to maintain that the mental – the domain of qualitative experience and understanding – has no counterpart on the micro-level.

— Joshua Rust on John Searle.


Cognitive science is increasingly interested in the embodiment of human perception, thinking, and action.

Proponents of this approach [hope it will dissolve] the Cartesian divide between the immaterial mind and the material existence of human beings (Damasio, 1994; Gallagher, 2005).

The shape, timing, and effects of [bodily actions such as pressing a button] are inseparable from their meaning.

— Georg Goldenberg, "How the Mind Moves the Body: Lessons From Apraxia" in Oxford Handbook of Human Action


The presumptions here are as follows.

The body is the entirety of an organism that is located spatially in a wider environment.

The brain is a body part specialised in monitoring what exists/happens in the environment, holding mental models and directing bodily actions.

The mind is that part of an organism that can form, remember and use bio/electro/chemically encoded forms of perceptions, memories, descriptions and directions.

The mind may primarily reside in the brain, but other parts of the body can be involved in mental processes.

Humans create and use concepts (encoded in the brain, on paper, or other) to describe and predict what exists/happens in that environment.

Where exactly the description of a concept is located inside the body is not important to the philosophy here.

Somehow, some way, the concept is encoded in the matter and energy contained within the body

And it doesn’t matter to us if that “somehow” proves forever obscure, incomprehensible and indescribable.



All free-to-read materials at are paid for out of income from Avancier’s training courses and methods licences.

If you find the web site helpful, please spread the word and link to in whichever social media you use.