Northoff Follow-up

I left out a few items, mainly of my own, on my last post either because they didn’t quite fit into the flow of the post or they didn’t occur to me at the time.

Consciousness Defined

Northoff defines consciousness as alignment between the world and the brain. This is brought about because the world is structured by temporal dynamics and the brain imitates these temporal dynamics in the actions of its neural circuits. Northoff starts with the world because the brain exists in the world.

What’s strikes me about this definition, when stripped of its technical jargon, is how much it agrees with what I call the commonsense definition of consciousness: the phenomenal states of being aware of the world usually accompanied by an ability to take actions in it. This certainly is what we do not have if we are in deep sleep, under anesthesia, in a coma, have fainted, or are suffering from a brain trauma, such as concussion. The questions asked of people in such states are: are they awake or are they conscious? This agrees with how Searle defines consciousness.

Consciousness in Other Organisms

Since consciousness is produced by the temporal dynamics of the brain mimicking those of the world, then presumably any organism with a brain that exhibits the scale free dynamics of networked neurons would have some level of consciousness.

The qualities and frequencies in which the brain aligns with the world, based on sensory capabilities, numbers of neurons, and complexity of brain structures, could differ remarkably across species n ways we may not be able to appreciate. The number of neurons, while perhaps not definitive, may be a useful proxy measure for level of consciousness. Number of neurons also maps to intelligence in some measures. While I don’t like to use the word “intelligence” as synonym for “level of consciousness”, I do think that level of consciousness mostly controls or determines intelligence in organisms with neural systems.

At any rate, in this argument, consciousness at some level likely arises with as few as a couple hundred to perhaps a thousand networked neurons. Some level of temporal dynamics (brain waves), I think, has been measured in insects.

Artificial Consciousness

If an artificial, non-biological creation could mimic the temporal dynamics of the world with its own internally generated temporal dynamics, we might be forced to conclude it is conscious. The created entity would likely require greater “sensory” access to the world possibly with actuators to enable movement than anything that exists today. One of the complaints about some of the AI models is that it can’t distinguish fact from fiction in what it is trained on. It has no contact or alignment with the world to verify what is and what isn’t.

What’s interesting about the human brain, however, is that it may have specialized circuits to tell us if what we are experiencing is real. There is a brain conditions called Capgras delusion, for example, that can make us think the people around us are not real. The anterior precuneus, a structure between the two hemispheres, seems to have some control over whether our individual reality feels real. So AI may need some kind of cross-check with itself to function better.

The question, of course, would be exactly what would be a minimal instantiation of the temporal dynamics. Would a lattice of silicon-based intelligent switches be sufficient? Or, is there something peculiar to calcium and potassium ions flowing in membranes, for example, that might require biology to be involved?

Thoughts on Time

We mostly think of time as either measured time or container time. In measured time, we are simply using some regular reoccurring event, for example, gear movement in a mechanical clock, as a standard for measuring some other event, for example the time from the start to finish of the 100 meter dash for a competitor. In container time, time moves at its pace and stuff happens in it. Container time is like older conception of space. In the Newtonian version, space isn’t anything but where matter exists. We know now matter bends space. Space is actively modified by matter.

What if time should be thought to be more active? The universe has matter but it also has the continual transformation of matter. Implicit in transformation is time. What was becomes something different through time. The world as we know has stuff (matter) but the stuff is always changing on short, long, and intermediate time scales. Time is active. It is stuff changing. Northoff calls this dynamic time and cites some of Lee Smolin to explain. The idea seems to be that the temporal dynamics has some creative role to play in how structures in the world come about. The same dynamics that build the world are used in the brain to create the structures that compose consciousness.

Posted in Friston, Time, Waves | 4 Comments

The Spontaneous Brain

Northoff, Georg. The spontaneous brain: From the mind-body to the world-brain problem. MIT Press, 2018.

Northoff, Georg. Neurowaves: Brain, Time, and Consciousness. McGill-Queen’s Press-MQUP, 2023.

The short story on both of these books is I like them. The first is complicated and challenging. The second is shorter and more accessible. They cover a lot of the same ground but there is less technical language challenges in the second. I consider Northoff an important voice and somewhat overlooked scientist and philosopher.

Initially I bought the Spontaneous Brain (SB) as an Amazon recommendation. That was a few years ago. When it arrived, I glanced at it and was somewhat daunted about the large amount of content. The book is dense reading, although maybe the problem in part is my getting older, more impatient – you know grumpy old man. It didn’t take long for something else to capture my attention and the book was put away on the shelf. A few weeks ago, while looking for a different book, I happened to spot the book again on the shelf and decided to pull it down for a second look. That look included the video I posted about previously.

SB is over 400 pages with dozens of pages of references and a glossary. The index has over 10 pages referencing Kant to give an idea about the scope of the book. Georg Northoff has a background in medicine, psychiatry, neuroscience, and philosophy and the book reflects a deep background in all of those things.

The second book, Neurowaves (NW), is recent and I found it when I began searching the Internet on Northoff. It is only around a hundred pages divided into six relatively short chapters. It covers much the same ground as SB but is easier to read. It is a more condensed view, although it is not “dumbed down” either.

Northoff is sometimes identified as a neurophilosopher. The book is a mixture of philosophy or neuroscience. The first two parts of SB are more neuroscience related and the next sections are more philosophical, but the book goes back and forth with its argument and blends it all into coherent whole. Of course, this makes the book difficult for some readers, myself included, who must be prepared to understand the terminology and concepts of two separate disciplines.

Northoff’s goal with these books seems broader than simply explaining the brain and consciousness. Although he acknowledges we still have gaps in our knowledge, much of the philosophical problem of mind-body and its hard problem variation arises from the substance-based metaphysics passed down from Descartes. Northoff is a ontic structural realist which is an ontology based on relationships rather than substances or properties. Ontic structural realism has as its core the idea idea that the structure of physical reality is relational. There are numerous nuances and variations of structural realism.

Northoff regards the mind-body problem as misguided and he isn’t offering a new solution to it. Instead, he wants to move the discussion to the world-brain relation. He writes” “Empirical evidence suggests that the brain’s spontaneous activity and its spatiotemporal structure are central for aligning and integrating the brain within the world – the world-brain relation; hence, the main title of this book.” (p vii) The spontaneous neurological activity of the brain ties directly to the ontology argument of the world brain relation.

Let’s step back a little and get back to neuroscience.

Brain and Consciousness

Probably the first question we should answer is what is meant by the “spontaneous brain” and its activity?

Scientists have been studying the details of what the brain does for more than a century. Even with one of the earliest tools, the electroencephalograph developed by Han Berger in 1924, it was realized that the brain has internal electrical activity that cannot be related to any external stimulus or task related activity. As scientists focused more on stimulus or task related research, study of the brain’s steady background activity was neglected. Recently that neglect has begun to be rectified and increasingly studies are appearing using the latest tools, like fMRIs, that seek to understand the brain’s spontaneous activity.

Northoff compares this situation in neuroscience to the philosophical differences between David Hume and Immanuel Kant. Hume emphasized a passive model of mind where everything was largely determined by external stimuli. Kant argued for an active model that stresses the importance of the mind’s internally generated activity.

Of course, it is clear that the brain both has its own internal, spontaneous activity and it responds to stimuli and task needs. Northoff believes that a complete account of neural activity requires a active brain that generates its own activity but also responds to the external world. He refers to this as a spectrum model where various mixtures and balances of internally generated activity and stimuli induced activity explain the brain’s range of activity.

The internal activity, however, does occupy a special place in this model. A key proposal of Northoff is that for external stimuli to be processed it must interact with the resting state brain activity and its spatiotemporal structure. Northoff writes about the spatiotemporal organizations of this spontaneous activity:

The brain’s spontaneous activity show a sophisticated structure that operates across different frequencies from infraslow over slow and fast frequency ranges. Importantly, the neural activity in these frequencies show a nested, or fractal, organization: just like a smaller Russian doll nests with the the next larger and so forth, the lower power of a faster frequency nests within the slightly higher power of the next slower frequency and so forth. Such temporal nesting is described as “scale-free dynamic,” since the same temporal relation (of slower and fast frequency power) holds across different (i.e. slow and fast) timescales.

NW p. 14-15

Scale-free dynamics “are characterized by hierarchical self-similarities of patterns of synaptic connectivity and spatiotemporal neural activity, seen in power-law distributions of structural and functional parameters and in rapid state transitions between levels of the hierarchy.” The slowest frequencies of the brain – slow and infraslow – generate a structured pink noise-like signal that builds the foundation for self and consciousness.

A prime example of the brain’s spontaneous activity is the so-called default mode network (DMN), a term coined by Marcus Raichle in 2001. Essentially the brain is active at a high level all of the time while a person is awake. Its energy consumption barely changes whether the person is engaged in an intense mental task or at rest. The DMN seems to take a backseat when the brain is engaged in an external task but it doesn’t switch-off as initially thought. There has been an explosion of interest in the DMN in the last decade or so since it was realized the DMN connects many different parts of the brain, especially ones related to self. Northoff points out that perhaps because of the DMN position in the middle of brain it seems to interact with more circuits than the other circuits do with each other. At any rate, the DMN has a definite spatiotemporal structure. It not only connects many regions of the brain, it also is integrated within its temporal wave patterns.

The DMN isn’t the only type of resting, self-initiated brain activity. It seems, in fact, that DMN as it is usually defined (unfocused brain activity) is more a subset of spontaneous activity that goes on all of the time in the brain. Spontaneous state activity is where most of the brain’s energy goes. Much of it occurs in the physical position of the middle of the brain. This not only gives it wide access but also provides a centering point, not left or right, not up or down, but physically aligned with the body. What’s more the range of frequencies encompasses both the fast (milliseconds) to very slow (many seconds/minutes). This suggests that consciousness doesn’t jump from instant to instant but smoothly changes over minutes and hours with spikes of activities from external stimuli that become assimilated into the core resting state activity.

There is a parallel in Northoff’s theory with global workplace theory which proposes that information from the senses must be broadcast throughout the brain to become conscious. New information from external stimuli is constantly brought into the resting state activity through wave propagation and interference patterns across the spatiotemporal structures of the brain. There are also parallels with aspects of Integrated Information Theory in that the primary function of this activity is integration of information from across the brain.

Northoff’s theory seems somewhat unique in that it doesn’t seem to provide some privileged position to a particular part of the brain, for example the frontal cortex. Consciousness isn’t created by a part of the brain but by the ongoing spatiotemporal dynamics of synchronized activity across the many parts of the brain and across time.

World Brain

First, let me note, it is world-brain not brain-world. Northoff believes the brain cannot be understood apart from the dynamics of the world in which it exists. The ongoing spatiotemporal alignment of brain with the world is at the core of what consciousness does.

Northoff especially points to research showing how the brain aligns itself with heart and the center of the body. We know also that brains begin to work differently (abnormally?) in sensory deprivation or outside of regular social contact. The brain regularly needs to resynchronize with the world in time and space. Depression, manias, and schizophrenia are characterized by poor temporal alignment between the brain and the world. People sharing the same activities synchronize their brain waves.

This world-brain alignment means that our consciousness bears some relationship to objective reality – that consciousness is not a total fantasy, hallucination, or illusion albeit there may be some “artistic” license in consciousness’s portrayal of that reality. Northoff explains:

Given these and other examples in nature like sea waves, wind, and bird-songs, scale-free activity is a universal feature in the world. The world constructs its own inner time in a scale-free way with temporal nestedness between the fluctuations of the different frequencies. The same hold true for the brain. As the brain is part of the world as a whole and its scale-free activity, the brain also constructs its own inner time in a scale-free way.

NW p 17

In other words, the brain exhibits scale-free temporal dynamics because it is a part of the world which exhibits the same dynamics. The brain and consciousness fractally represent the world by using the same temporal dynamics.

Northoff believes that shifting the focus to world-brain will create an equivalent of a Copernican revolution in neuroscience and philosophy.

My Thoughts

All in all, I still not sure I understand all of Northoff’s positions or how they all fit together. Hopefully nothing significant that I’ve written is wrong or misleading. At any rate, I have tried to provide a few snapshots into some of Northoff’s arguments but let me be clear, there is a lot of his arguments that I have left out.

My initial interest in Northoff’s ideas came from the emphasis on the explanatory power of spatiotemporal structure for explaining the brain. After reading Pockett and McFadden on EM field theories and their emphasis on the spatiotemporal aspects of the brain, I had begun to suspect that the physical structure of neurons and their relationships – the actual geometrical architecture – combined with temporal variations might be a common language between the physical brain and the elements of consciousness. If this is correct, eventually some day we may understand the phonemes and sentences of the language.

To fully explain consciousness we may still need more than a common language. Some candidates would include EM field, extra spacetime dimensions, a quantum element, or a new type of wave, field, or particle (psychons, anyone?). Northoff, I think, has made a big step in pointing out the importance of spatiotemporal architecture without the complications involving additional explanatory elements.

That AI trained on fMRI data can begin to “mind read” even to a limited degree suggests that spatiotemporal patterns themselves must bear some relationship to the associated phenomenal events. Of course, the patterns could be correlated but epiphenomena. What argues against that in my mind is that geometric behavior exists in many other natural forms. Life itself is primarily based on L- varieties of sugars and amino acids rather than the twin D- varieties. This is essentially a geometric difference in how the molecule forms. DNA, RNA, proteins, and so on, almost of the chemicals of life have complex and dynamic geometric (and electromagnetic) structures. In many cases, the geometrical shapes of the molecules, how they fit together, is key to how they interact or fail to interact. Scale-free dynamics appears in nature, in ocean waves, wind, and bird songs for example, which suggests that it may be core organizing principle in complex phenomena. That the human brain and the consciousness it generates would have characteristics of natural phenomena, indeed, reflects its fractal nature.

Northoff’s notion of the world-brain didn’t grab me at first. Tentatively, before reading Northoff, I had already begun exploring the idea that consciousness may be more public than we might think. We have common shared metrics. We can agree on how to measure distance, time, and other physical relationships. We can even agree when we both see blue (most of the time). Certainly we also share a common substrate. Although each of our brains may have an architecture as individual as our fingerprints, in major ways normal, uninjured brains look, behave, and, from most reports, work similarly. Not surprisingly when people engage in common activities like singing, dancing, working on a group task, their brain waves also begin to synchronize. Those are social settings, of course, but there is no reason that principle doesn’t extend from the brain to the social and natural world as Northoff suggests.

It isn’t hard to see how the fractal relationship between world and brain would come about, even almost be compelled by the mimetic nature of networked neurons. If neurons imitate the world, then the most important part of that imitation would be maintaining the same or similar relationships between what exists in the world and what exists in neuronal relationships. If the world is based on temporal scale-free dynamics, then the brain would have to adopt the same to represent it.

The world-brain relation and its alignment could be exactly what is at the root of much consciousness alteration efforts. The therapeutic effect of psychedelics derives from the opportunity to rewire connections or, in Northoff’s terminology, bring about the alignment of brain, body and world. Would not the ultimate alignment of brain and world be cosmic consciousness? Perhaps it isn’t so strange that spiritual traditions, especially in the East but also in the esoteric branches of Western religions, have embraced physical practices that deliberately or inadvertently may be effective techniques for modifying brain wave patterns to more fully align with the world. In addition to the breathing and postural techniques of yoga and similar practices, we could add chanting, prayer, and mantra reciting. What the shaman and his techniques attempt to cure is a misalignment with the world. What the shaman promotes is the maintaining of existing harmony and a realignment when harmony is disrupted. The ultimate end for alignment would be a consciousness fractally representing the world in a universal chorus of conscious beings.

Posted in Consciousness, Entropy, Friston, Information | 24 Comments

Consciousness as Mimicry

I’ve compared the mind to a model but I’ve struggled to find a good example of the type of model it is.

We know there are different types of models. In science most models are abstract and mathematical. A couple of equations with tunable parameters could be used to model a complex physical interaction. For example, a climate model might need to account for variations in the Earth’s orbit and spin, influence of volcanic eruptions, solar variations, geological and oceanic changes, the effect of greenhouse gases, and the efficiency of carbon sinks. In a perfect model, the known and best estimate variables would predict accurately the variations in the temperature and ice caps over millions of years. Another example of this same type of model would be using physics equations to graph the arc of a projectile fired at specified velocity.

While the mind might be modelled abstractly, the mind itself isn’t an abstract model. Mathematical models quickly break down in modeling complex interactions. Even models of the interactions of as little as a hundred particles would require the computing power of the universe. While abstract models can be dynamic and arrive at different results with different inputs, they are frequently narrow in scope.

In contrast to abstract, mathematical models, there are concrete, physical models. The simplest of these to understand would be the scale model. Some of the earliest scale models were of buildings and dwellings from several thousand years BCE. Scale models are still extensively used in architecture, but the use has expanded to cars, rockets, and action heroes.

My first attempt at describing mind as model was the airplane in the wind tunnel. The problem with the example is that it is a static model and the mind is not static. We can put a model airplane in a wind tunnel and measure turbulence and lift, but the model just sits there. It doesn’t change. We can pull it out of the wind tunnel and change it, but it does not change itself. A second attempt was the YouTube video of kinetic art. This model isn’t static but it is very mechanistic and deterministic.

I was watching My Octopus Teacher for the third time with my wife (she hadn’t seen it) a few days ago. The movie is on Netflix and tells the story of the interaction between a man and an octopus over the course of a year. It begins with the octopus camouflaged in shells. Later we learn that the octopus does this to hide and protect itself from sharks. Through the movie, the octopus hides in kelp, merges with the sea bottom, and adopts the colors and textures of its background. Apparently the mimic octopus has even greater capabilities and can emulate more than a dozen creatures to trick predators and prey.

Mimicry in nature usually arises slowly through evolution and typically involves a harmless organism taking on the appearance or characteristics of a dangerous or poisonous organisms to fool predators. This is usually static. Once the organism develops it can’t change its appearance The dynamic sort of mimicry found in chameleons and octopi is more rare. In the octopus, apparently the skin has thousands of are chromatophores controlled by nerves that the octopus can squeeze or stretch to change their reflective properties.

The mimicry of the octopus is like the brain/mind model. First, it is biological. It is also dynamic. It can change rapidly based on environment. It is, I suppose, somewhat involuntary. The octopus may decide to take a form, color, and texture but we can suppose that whatever mechanism it uses to control each chromatophore is unconscious and automatic. Most importantly, however, the mimicry of the octopus reflects the world. In fact, it has been used to argue against extreme “no objective reality” views such as Donald Hoffman’s. Applied to mind and consciousness it could also be used to argue for some fundamental veridicality of mind.

This doesn’t mean necessarily that the mimicry of the octopus works identically to how the brain forms its model of the world. I am only arguing for similarity and likeness. The characteristics of mimicry – physicality, living, dynamic reaction to environment, reflective of reality – are similar.

Mind is essentially an internalized mimicry of the external world. The way it mimics is by creating spacetime relationships between sense datum coming from the senses. Closely related data become objects. The temporal-spatial relationships of objects is mimicked by neuronal speed (time) and the dimensional and patterns in the spatial architecture of of neural connections. This mimicry is what I have been calling a model and it maps (probably polymorphically) its internal objects with external objects based on a best fit for the sense data available.

If mind is internalized mimicry, the problem of explaining how it works becomes drastically simplified. We don’t have to imagine how complex computational algorithms evolved and became embedded in the brain. We just have to assume that the primary function of the brain is to mimic external reality by maintaining internally the same relationships that the sense data presents. The brain and its neurons doesn’t to know anything. It just has to faithfully copy the relationships and revise its copy whenever new data appears.

There certainly is more than this but this gets directly at how neurons can seemingly know about the world.

Posted in Brain size, Consciousness, Information, Intelligence | 63 Comments