Scientific American article on IIT

This article is quite old (2015) but it is interesting because it questions Integrated Information Theory. The author is John Horgan of end of science fame.

https://blogs.scientificamerican.com/cro...ciousness/

I didn't even know Scientific American had blogs!

Anyway, it is definitely worth reading, even without understanding IIT - which seems remarkably dense.

Also it would seem that someone has devised a simple 2D grid of XOR gates that could have arbitrarily high phi - and is being touted as a counter example to the whole theory.

David

Dense language, for sure.  Here is Tononi & Koch trying to describe the process of mind, imagined from the perspective of physics.

According to IIT, the quantity and quality of an experience are an intrinsic, fundamental property of a complex of mechanisms in a state—the property of informing or shaping the space of possibilities (past and future states) in a particular way, just as it is considered to be intrinsic to a mass to bend space–time around it.

I would love to talk about just this in depth.  It is on the right tracks, in many ways, to my perspective.

Does it communicate anything with its reaching out to Einstein's ideas?

Yes, it sounds rather like hunting for prime numbers, or something rather more exotic and therefore rarer. However anything like that isn't physical at all - it is an idea that exists for ever.

Here you will find a list of axioms for IIT:

https://en.wikipedia.org/wiki/Integrated...ion_theory

They seem rather odd to be axioms for anything - what do you think?

My thoughts are Tononi and Sporns see themselves as following the science of the day.  Here is background that may be useful in following how ITT got its legs.  They set out, like all good scientists to generate a process map. One that maps the brain's abilities, which have defined variables and measures.  The axioms present that work and do address the processes of thinking.  

On the personal side - as usual - I point to primacy of the mental process of understanding.  It is expressed by ITT as integration.

Sporns is not as well known.

 Sporns’ main research area is theoretical and computational neuroscience, with an emphasis on complex systems, brain connectivity, and neurorobotics. His work aims to uncover the network principles that underlie the architecture and function of the human brain. He is particularly interested in how anatomical connections shape and constrain functional brain dynamics, and how disruptions of anatomical connections may relate to neural and mental diseases. An essential step towards a better understanding of human brain networks is the creation of a complete connection map, which Sporns has called the “human connectome."

https://www.gf.org/fellows/all-fellows/olaf-sporns/

The paper that started it all.
https://bmcneurosci.biomedcentral.com/ar...-2202-4-31

 In this way we define the effective information from A to B as: where MI(A:B) = H(A) + H(B) - H(AB) stands for mutual information, the standard measure of the entropy or information shared between a source (A) and a target (B).

This is bedrock math and comes directly from Claude Shannon and the MTC.  It is the basic equation to compare before and after transmission.  They then link this informational measure in units of bits, to a logical framework, where the outcome of a signal (stimulus) is the variable.

... In neural terms, we try out all possible combinations of firing patterns as outputs from A, and establish how differentiated is the repertoire of firing patterns they produce in B. Thus, if the connections between A and B are strong and specialized, different outputs from A will produce different firing patterns in B, and EI(A→B) will be high. On the other hand, if the connections between A and B are such that different outputs from A produce scant effects, or if the effect is always the same, then EI(A→B) will be low or zero. ...

The grand equation -

 For a given bipartition [A:B]S of subset S, the effective information is the sum of the effective information for both directions:
 

All of the space and activity parsed by this equation is about variables of knowing, understanding and intent.  Not about an electro-chemical reaction.

My thoughts are Tononi and Sporns see themselves as following the science of the day.  Here is background that may be useful in following how ITT got its legs.  They set out, like all good scientists to generate a process map. One that maps the brain's abilities, which have defined variables and measures.  The axioms present that work and do address the processes of thinking.  

On the personal side - as usual - I point to primacy of the mental process of understanding.  It is expressed by ITT as integration.

Sporns is not as well known.
https://www.gf.org/fellows/all-fellows/olaf-sporns/

The paper that started it all.
https://bmcneurosci.biomedcentral.com/ar...-2202-4-31



This is bedrock math and comes directly from Claude Shannon and the MTC.  It is the basic equation to compare before and after transmission.  They then link this informational measure in units of bits, to a logical framework, where the outcome of a signal (stimulus) is the variable.

The grand equation -

All of the space and activity parsed by this equation is about variables of knowing, understanding and intent.  Not about an electro-chemical reaction.

"The science of the day" has proven to be totally inadequate in dealing with this problem. None of this math goes even an inch toward understanding how subjective experience and qualia and knowing and willing and all the rest of the qualities and attributes of consciousness emerge from equations or the variable neurological measurable parameters they are based on. Neither IIT nor other approaches based on information theory or physics seem to really go anywhere in dealing with the Hard Problem. That's why it's so hard.

My thoughts are Tononi and Sporns see themselves as following the science of the day.  Here is background that may be useful in following how ITT got its legs.  They set out, like all good scientists to generate a process map. One that maps the brain's abilities, which have defined variables and measures.  The axioms present that work and do address the processes of thinking.  

On the personal side - as usual - I point to primacy of the mental process of understanding.  It is expressed by ITT as integration.

Sporns is not as well known.
https://www.gf.org/fellows/all-fellows/olaf-sporns/

The paper that started it all.
https://bmcneurosci.biomedcentral.com/ar...-2202-4-31



This is bedrock math and comes directly from Claude Shannon and the MTC.  It is the basic equation to compare before and after transmission.  They then link this informational measure in units of bits, to a logical framework, where the outcome of a signal (stimulus) is the variable.

The grand equation -

All of the space and activity parsed by this equation is about variables of knowing, understanding and intent.  Not about an electro-chemical reaction.

Well Stephen,

If you think you know how any of this works (I confess I don't), select whatever version of those axioms you prefer, plus some simple physical system from some of Tononi's writings, and let's see if we can precisely figure out what it means - even calculate its phi - collectively on this forum.

Even the example in Koch's book - a triangle of three gates - and OR gate, an XOR gate and a 'copy gate' (which I assume copies its input at time t to its output at time t+1) isn't finished - he doesn't show and answer like phi=<something>.

A properly worked example might really cut through the confusion. I'd even like to know what units phi is measured in, and what range of values is possible (0 - 1, 0 to infinity, or what).

After that I'd like to think about the 2-d array of XOR gates, that is reputed to have unbounded phi as the size of the array grows.

That sounds somewhat like the old 'game of life' construct.

David

"The science of the day" has proven to be totally inadequate in dealing with this problem. None of this math goes even an inch toward understanding how subjective experience and qualia and knowing and willing and all the rest of the qualities and attributes of consciousness emerge from equations or the variable neurological measurable parameters they are based on. Neither IIT nor other approaches based on information theory or physics seem to really go anywhere in dealing with the Hard Problem. That's why it's so hard.

I am not so afraid of the Prof Chalmers framework.  It is intractable from the point of "meaning" being a product  of equations with only units of measure from materials science and physics (SI units).

I am arguing for "meaning" to be substance, which is quantifiable as to its structure and quantifiable as to the logic of its patterned  activity.

There is another set of units that information and communication science is building.  These units of measure address immaterial aspects of reality and enable quantification of activities like decision-making and ecological adaptation.  The space of this ecology is not the space of 3 dimensions.  There are multitudes of informational spaces, where activity is graded by outcome and logic.  In this environment - life thrives by the outcomes influenced by mind.  The models of the transformation of entropy to negentropy, through mind are becoming clear.

The Hard Problem's ground and arguments change completely when information activity is seen as having independent variables from physics (information objects) and that real-world probabilities are quantifiable as substance that is NOT MIND INDEPENDENT.

Therefore, whether or not there is a subjective physicalist bend to the data analysis - Sporns, Tononi, Koch et all are gathering data from informational environments.  This is the data I find exciting, because it documents mental activity.  I am so confident that the better we know physics - the more clear its relationship to spirit/mind is defined as mutual, but separate.

I am not so afraid of the Prof Chalmers framework.  It is intractable from the point of "meaning" being a product  of equations with only units of measure from materials science and physics (SI units).

I am arguing for "meaning" to be substance, which is quantifiable as to its structure and quantifiable as to the logic of its patterned  activity.

There is another set of units that information and communication science is building.  These units of measure address immaterial aspects of reality and enable quantification of activities like decision-making and ecological adaptation.  The space of this ecology is not the space of 3 dimensions.  There are multitudes of informational spaces, where activity is graded by outcome and logic.  In this environment - life thrives by the outcomes influenced by mind.  The models of the transformation of entropy to negentropy, through mind are becoming clear.

The Hard Problem's ground and arguments change completely when information activity is seen as having independent variables from physics (information objects) and that real-world probabilities are quantifiable as substance that is NOT MIND INDEPENDENT.

Therefore, whether or not there is a subjective physicalist bend to the data analysis - Sporns, Tononi, Koch et all are gathering data from informational environments.  This is the data I find exciting, because it documents mental activity.  I am so confident that the better we know physics - the more clear its relationship to spirit/mind is defined as mutual, but separate.

A famous (in philosophy of mind circles) thought experiment proposed by Frank Jackson encapsulates the fundamental problem posed by the Hard Problem. This is the Mary's Room thought experiment.

It was originally proposed by Frank Jackson as follows (Wiki):

"Mary is a brilliant scientist who is, for whatever reason, forced to investigate the world from a black and white room via a black and white television monitor. She specializes in the neurophysiology of vision and acquires, let us suppose, all the physical information there is to obtain about what goes on when we see ripe tomatoes, or the sky, and use terms like "red", "blue", and so on. She discovers, for example, just which wavelength combinations from the sky stimulate the retina, and exactly how this produces via the central nervous system the contraction of the vocal cords and expulsion of air from the lungs that results in the uttering of the sentence "The sky is blue". ... What will happen when Mary is released from her black and white room or is given a color television monitor? Will she learn anything or not?
...................................................
- Mary (before her release) knows everything physical there is to know about other people.
- Mary (before her release) does not know everything there is to know about other people (because she learns something about them on her release).
- Therefore, there are truths about other people (and herself) which escape the physicalist story."

These truths are the fundamentally immaterial essences of subjective consciousness such as qualia, thought, intentionality, etc. They fundamentally "escape the physicalist story" where this story includes theories based on information theory and physics and math such as IIT. This includes the "Units of measure" in the IIT theory which symbolize and attempt to objectify these immaterial properties of consciousness. It's the same as Mary's attempt to objectify and measure the essence of the immaterial qualia of subjective color perception. Can't be done.

A famous (in philosophy of mind circles) thought experiment proposed by Frank Jackson encapsulates the fundamental problem posed by the Hard Problem. This is the Mary's Room thought experiment.

It was originally proposed by Frank Jackson as follows (Wiki):


These truths are the fundamentally immaterial essences of subjective consciousness such as qualia, thought, intentionality, etc. They fundamentally "escape the physicalist story" where this story includes theories based on information theory and physics and math such as IIT. This includes the "Units of measure" in the IIT theory which symbolize and attempt to objectify these immaterial properties of consciousness. It's the same as Mary's attempt to objectify and measure the essence of the immaterial qualia of subjective color perception. Can't be done.

Physical information????  What do you think this excludes?  By definition the choices of living things that are possible - but we don't make - count for nothing.  This is strongly disputed.

Is information theory part of the Physicalist story!?   I need to think about how that may be true, while seeing it as a Trojan Horse to their cultural memes.

The informational realist position (my version only) says Mary gains information.  The colors may/should carry meaning for her experiences of color that occur.  It may have connection to other objects in memory that had that wavelength.  And at a deeper level, feelings from the subconscious experience of collective knowledge from the information environment.
(think Jung)

Tononi et all, aren't trying to objectify "immaterial properties".  I suggest he cares little.  He is trying to have objective data about neural connections and behavioral outcomes.  The "properties" of experience and the existence of mental tasks are described.  And described well.

 I would simply have a different perspective, that clearly divides physical steps of transformation, from mental ones acting directly of the real-world probabilities.

Well Stephen,

If you think you know how any of this works (I confess I don't), select whatever version of those axioms you prefer, plus some simple physical system from some of Tononi's writings, and let's see if we can precisely figure out what it means - even calculate its phi - collectively on this forum.

Even the example in Koch's book - a triangle of three gates - and OR gate, an XOR gate and a 'copy gate' (which I assume copies its input at time t to its output at time t+1) isn't finished - he doesn't show and answer like phi=<something>.

A properly worked example might really cut through the confusion. I'd even like to know what units phi is measured in, and what range of values is possible (0 - 1, 0 to infinity, or what).

After that I'd like to think about the 2-d array of XOR gates, that is reputed to have unbounded phi as the size of the array grows.

That sounds somewhat like the old 'game of life' construct.

David

I am not able to take on the task you describe, but still think that I can add to the general picture of ITT, if we pull it apart ourselves.

http://www.scholarpedia.org/article/Inte...ion_theory

 The system must specify a cause-effect structure that is the particular way it is: a specific set of specific cause-effect repertoires—thereby differing from other possible ones (differentiation). A cause-effect repertoire characterizes in full the cause-effect power of a mechanism within a system by making explicit all its cause-effect properties. It can be determined by perturbing the system in all possible ways to assess how a mechanism in its present state makes a difference to the probability of the past and future states of the system. Together, the cause-effect repertoires specified by each composition of elements within a system specify a cause-effect structure. Consider for example, within the system ABC in Figure 3, the mechanism implemented by element C, an XOR gate with two inputs (A and B) and two outputs (the OR gate A and the AND gate B). If C is OFF, its cause repertoire specifies that, at the previous time step, A and B must have been either in the state OFF,OFF or in the state ON,ON, rather than in the other two possible states (OFF,ON; ON,OFF); and its effect repertoire specifies that at the next time step B will have to be OFF, rather than ON. Its cause-effect repertoire is specific: it would be different if the state of C were different (ON), or if C were a different mechanism (say, an AND gate). Similar considerations apply to every other mechanism of the system, implemented by different compositions of elements. Thus, the cause-effect repertoire specifies the full cause-effect power of a mechanism in a particular state, and the cause-effect structure specifies the full cause-effect power of all the mechanisms composed by a system of elements.[8] 

This is the information that is going to populate the fields in the database.  It is seeing our minds like a circuit board.  What is connected to what, with the strength of the outcome figured in.

Physical information????  What do you think this excludes?  By definition the choices of living things that are possible - but we don't make - count for nothing.  This is strongly disputed.

Is information theory part of the Physicalist story!?   I need to think about how that may be true, while seeing it as a Trojan Horse to their cultural memes.

The informational realist position (my version only) says Mary gains information.  The colors may/should carry meaning for her experiences of color that occur.  It may have connection to other objects in memory that had that wavelength.  And at a deeper level, feelings from the subconscious experience of collective knowledge from the information environment.
(think Jung)

Tononi et all, aren't trying to objectify "immaterial properties".  I suggest he cares little.  He is trying to have objective data about neural connections and behavioral outcomes.  The "properties" of experience and the existence of mental tasks are described.  And described well.

 I would simply have a different perspective, that clearly divides physical steps of transformation, from mental ones acting directly of the real-world probabilities.

Your information realist position as you describe it says nothing to explain Mary's conscious subjective experience of color once she is released. Neither does Tononi's IIT as you describe it. Therefore these two positions or theories do not address the central mystery of consciousness, which is what I thought this was all about. It looks like there must be a lot of hype involved in promoting the IIT theory, essentially claiming that it does far more than it really covers.

Stephen,

Here is a good (but sceptical) article about IIT by Scott Aaronson:

https://www.scottaaronson.com/blog/?p=1799

It is partly mathematical, but the maths seems more precise than in other places, and I'd judge it might be doable, except that having read his conclusions I wouldn't have much motivation to make the effort!

Part of his objection seems to be that he can construct examples of systems with high phi but it is implausible that they are conscious, but his philosophical points are interesting too:

The most obvious thing a consciousness theory could do is to explain why consciousness exists: that is, to solve what David Chalmers calls the “Hard Problem,” by telling us how a clump of neurons is able to give rise to the taste of strawberries, the redness of red … you know, all that ineffable first-persony stuff. Alas, there’s a strong argument—one that I, personally, find completely convincing—why that’s too much to ask of any scientific theory. Namely, no matter what the third-person facts were, one could always imagine a universe consistent with those facts in which no one “really” experienced anything. So for example, if someone claims that integrated information “explains” why consciousness exists—nope, sorry! I’ve just conjured into my imagination beings whose Φ-values are a thousand, nay a trillion times larger than humans’, yet who are also philosophical zombies: entities that there’s nothing that it’s like to be. Granted, maybe such zombies can’t exist in the actual world: maybe, if you tried to create one, God would notice its large Φ-value and generously bequeath it a soul. But if so, then that’s a further fact about our world, a fact that manifestly couldn’t be deduced from the properties of Φ alone. Notice that the details of Φ are completely irrelevant to the argument.

Faced with this point, many scientifically-minded people start yelling and throwing things. They say that “zombies” and so forth are empty metaphysics, and that our only hope of learning about consciousness is to engage with actual facts about the brain. And that’s a perfectly reasonable position! As far as I’m concerned, you absolutely have the option of dismissing Chalmers’ Hard Problem as a navel-gazing distraction from the real work of neuroscience. The one thing you can’t do is have it both ways: that is, you can’t say both that the Hard Problem is meaningless, and that progress in neuroscience will soon solve the problem if it hasn’t already. You can’t maintain simultaneously that

(a) once you account for someone’s observed behavior and the details of their brain organization, there’s nothing further about consciousness to be explained, and

(b) remarkably, the XYZ theory of consciousness can explain the “nothing further” (e.g., by reducing it to integrated information processing), or might be on the verge of doing so.

Read the non-mathematical parts, and then maybe discuss the subject.