Is mental representation innate?

My impression is that the proposed innateness of mental representation is the logical entailment of traditional computer-science-inspired information-processing theories of cognition. Fodor famously advertised this option as the only choice left open to cognitive science as it has so far expressed itself. The fact that it has been his reductio-ad-absurdum joke on cognitive science has been lost on many people who either took Fodor seriously to mean that cognitive structure is in fact innate or read him incompletely enough to mistake punchline for premise.

Also, regarding the question of whether anything is evidence for innateness, I recommend reading the work in developmental science by Gilbert Gottlieb and his students on the nonobvious experiential effects on seemingly innate behaviors.

"proposed innateness of mental representation is the logical entailment of traditional computer-science-inspired information-processing theories of cognition"

I'm with you on that, though representationalism goes back to Descartes.

" fact that it has been his reductio-ad-absurdum joke on cognitive science has been lost on many people"

nicely put. Though I think Fodor went along for the ride.

"Gilbert Gottlieb"

will check

thanks

SP

Indeed, Fodor quips that Descartes started the computer metaphor for cognition 400 years before computers were invented. 

Of course, if we may be as cutesy as Fodor with our quips, then we might remember that before we settled on the marketable model of the computer, smart people were playing with the possibility that animal swarms (e.g., Daphnia water flies) could "compute" complex functions too.  Daphnia emerged on the evolutionary scene quite a bit before our desk ornaments did. :)

See attached work by Peter Cariani on computation and emergence.

It seems that our capacity for representation is innate as is evidenced by differential neonatal responses to prenatal exposures to music, speech rhythms and tones of mother, etc. Thus while the capacity for mental representation is innate, mental contents are acquired.

Linda "Thus while the capacity for mental representation is innate, mental contents are acquired."

Yes, indeed. If we each did not have the innate brain mechanisms which enable us to represent the volumetric space around us (I! -- our own unique locus of perspectival origin), we would have no conscious experience of the world we live in, and our exchange of opinions on RG could not possibly take place. For more about this, see "Where Am I? Redux" and "Evolution's Gift: Subjectivity and the Phenomenal World" on my RG page.

What evidence do we have of internal representations at all?

Thankyou to all who have replied. I suppose I inadvertently laid a trap by posing the question in a way which would divide responses. Clearly hard line responses on either side of the nature/nurture biology/culture dual will be difficult to support , a 'both-and' approach seems sensible. 

I agree with Gabriele that a large amount of the subtlety of our representational capability must be learned- when I recognise a waltz, or a biblical quotation, I am referring to cultural entities. But when I dance a waltz, what role do representations play? When I pick up the scent of frangipani, do I need a mental model of frangipan scent in order to recognsie it?

Serban asks a key question, and as with AI debates of 25 years ago, there seems to be no agreement on what a representation is,  how a representation might be constituted or where we might find one. By introspection we believe we have them- I can call the visage of a friend to mind. Yet how that mental image is prepared in order for us to have the impression that it is presented to the 'mind's eye' is afaik, unknown.

These are higher order conscious representations. What about hunger or the sense of falling - do these require representations for us to know them?

Representations are distinct patterns of neuronal excitation in the brain that are systematically evoked by real objects and events, or by imagined objects and events. Such neuronal representations provide the biological grounding for the semantics of our languages. See, for example, Linking the Semantic Network to the Real World in "Building a Semantic Network" on my RG page.

Response to Arnold Trehub

Dear Arnold,

At your suggestion I reviewed a couple of your papers, specifically Space, Self and the Theatre of Consciousness, and Evolution’s Gift, as well a your response to Noe. A set of traditional ideas which are quite familiar to anyone who has read philosophy of mind and cognitive science underpin your positions generally and particularly your Retinoid System hypothesis. Your reference to a theatre of consciousness clearly recalls the Cartesian Theatre, and this brings with it most of the other ideas I have trouble with.

In the Theatre of Consciousness paper you refer to “The need for an egocentric representation of 3D space”. I query that assertion. Why do we need a singular unified representation of (egocentric) space? I think we have multiple overlapping but not necessarily congruent senses of space conferred by different senses, or combination of senses.

The temporal dynamic aspect of ‘sense of place’ seems to take a back-seat to a static, image based approach. I would argue that our sense of being in the world is conferred upon us largely by dynamical qualities involving proprioception and kinesthesia. Neuroscience seems to bear this out in various ways. ie evolutionarily, the brain evolved first to coordinate motion. And movement in the world is a matter of sensori-motor feedback loops.

As per the Cartesian Theatre, your Retinoid System subscribes to the enlightenment

Idea of the primacy of vision. Yet neurologically, large-scale geographical location seems to be a matter for grid and place cells, far from visual centers. How do you accommodate this research with your supposition of the location of 3D spatiality in the hypothetical Retinoid System?

Another Cartesian quality in your writing is the extensive deployment of quadrilateral geometries. But there isn't a right angle in the body, nor was there in the world before humans created them. Yes, I admit, we tend to be perpendicular with respect to an idealised ground plane, but that perpendicularity does not imply a world xyz space. Interestingly, grid cells are by no means xy, but seems to work on a kind of moiré system of resonances around a hexagonal geometry.

The grid and place cells seem to operate on a geographical scale, far beyond the range of effective stereoscopy, which is really effective only for hand/eye coordination (which is where we need it) and somewhat useful for close bodily motion, ie descending stairs. I would say that your 3D egocentric space, is the case (only) within the zone of stereoscopy and hand eye coordination. We are peripersonally 3D.

In terms of larger geographical space, I would say we are ~2.1 dimensional. I mean this in the fractal sense, and it is born out by neurological research. For instance, the horizontal planar grid cell operation seems to be anchored by gravity. It does not reorient to the vertical plane if a person, or a rat, is confronted with a vertical landscape. We simply don’t do the vertical axis very well, which is partly why our architecture is largely made up of layered horizontal planes, as opposed, say to intersecting oblique planes.

In your response to Noe, you say “The world that you (or I) consciously experience (the

phenomenal world) is, in its irreducible form, a global coherent volumetric space in which you are the egocentric origin of all experience and action.”

I wonder why you insist on the need for a unified or coherent personal space. I imagine we can get along quite well without it. Nor do I assume a unified self, for that matter. For me, this desire for coherence echoes the late modern yearning for unified everything – Principia Mathematica, grand unified field theory, etc.

But the world isn’t like that, Biology and evolution are full of inconsistencies, happy accidents of neural exploitation, etc. We demonstrably have different ie sensory system, and sometimes they disagree –in motion sickness, and in simulator sickness, for instance.

I see no reason to assume that cognition is (always or even mostly) a matter of reference to internal representation, even if it is a global spatial representation. The notion of the image, so central to enlightenment thought, has the qualities of being static and being representational, and privileges vision. According to my research, many of these assumptions which undergird traditional cognitive science just don’t ‘map’ onto what we are learning about the brain, and so in my opinion, they just get in the way of building a workable set of theories about cognition.

Our philosophical traditions, in the main, struggle with process and the processual. But I think this is cultural. We like to think of spatiality but not of temporality. That is why we think it terms of (static) representation as opposed to the idea that cognition is a constant dynamical performance of negotiating the world we experience.

Damian

thanks for the Cariani ref. I've known him for years and admire, and often quote, his work

Simon,

I appreciate your detailed response. You wrote "Our philosophical traditions, in the main, struggle with process and the processual. But I think this is cultural. We like to think of spatiality but not of temporality. That is why we think it terms of (static) representation as opposed to the idea that cognition is a constant dynamical performance of negotiating the world we experience."

Your suggestion that cognition is "a constant dynamical performance of negotiating the world we experience" is on the mark, but I believe you are mistaken to think that representation in retinoid space -- our 'theater of consciousness' -- is only visual and static. It is always dynamic and normally multi-sensory. A distinction must be made between the immediate global unity of the egocentric conscious presence and our more fragmented perceptions and thoughts that are the result of darting foci of attention within our phenomenal world, and our efforts at cognitive analysis. Grid cells and place cells are part of the machinery of spatial cognition but they do not represent our enveloping phenomenal world. They are learned neuronal tokens that guide behavior. And it is unlikely that the severely locked-in author Jean-Dominique Bauby's vivid representations of being in the world were determined by proprioceptive and kinesthetic signals (see The Diving bell and the Butterfly. Vintage 1998).

This is an important topic that is worth exploring deeper. What is the extended present?

The concept "innate" has no defensible meaning within the modern sciences. You might choose to adopt some practical (but inevitably somewhat arbitrary and idiosyncratic) definition, such as "is very likely to develop fairly early in life" or "present at birth" (though we have a more precise term for that, i.e., congenital). But the last 50 years of research in developmental biology (among other fields) more or less force a reconceptualization of trait development and gene activity within which the folk-biological concept 'innate' has no analytically defensible definition. Put more simply, there is no way to apply the description "innate" to any trait, even with qualifiers, that would clearly communicate some quality of that trait. The proper use of the word "innate" now lies outside of the sciences, as a metaphor in everyday language (e.g., "The spirit moved me"; "She's innately good at math"; "Wow, the gods must be smiling on you!").

Gedeon

thankyou for your succinct and thought provoking response - which takes the question in a direction away from representationalism to the matter of traits. For my own clarification, and as I want to get my terminology right within your terms of reference:  - I have characteristics in common with most of my conspecifics - opposable thumb, binocular vision. These, I assume, are not 'traits' but 'congenital' (normalities?). Or is binocularity a trait, whereas having two eyes with the potential of stereoscopy is a characteristic? 

Since you are clearly concerned with representation, how would you phrase my original question, or is it simply somehow wrongheaded?

Thankyou also for posting your paper Development of Adaptive

Tool-Use in Early Childhood, I've downloaded it as it speaks directly to some issues I am working on.

Simon

Amherst

Gedeon: "Put more simply, there is no way to apply the description "innate" to any trait, even with qualifiers, that would clearly communicate some quality of that trait."

Mental representation is not just a trait. Mental representations are biophysical events produced by particular kinds of innate brain mechanisms.

Hi Simon, thanks for an interesting conversation - and for downloading the paper. Hope you find it useful. I welcome any feedback.

To clarify: I'm not taking on the construct 'trait' (although it begs for deconstruction). To simplify we can ask: is it meaningful to describe some traits as more innate (e.g., thumbs) and some as less so (e.g., tends to score high on math tests)? If useful, in what sense -- pragmatically general, or actually carrying information? It does carry pragmatic information -- for example, like it or not, it implies something about the training/expertise of the person who uses the word regularly. Does it carry information relevant to scientists? Not much, because all traits, including anatomical traits, are emergent products of a history of exposure to environments that change the physiology and structure of the cells involved. Change the concentration of SSH, FGF4 or FGF10 or certain WNTs or RA or a number of other transcription factors on certain days (i.e., ranges of cell cycles) after fertilization, and you will change the structure of limb buds, or lung buds, or (surprise) cortical projection fibers. Some molecules serve as gene modulators prenatally and as neurotransmitters postnatally (and prenatally); the expression of many of these (e.g., cortisol; Ne) is affected by various factors including the local and more distal (extra-organism) environment. There are so many examples...using addictive drugs, e.g., cocaine/crack, causes local/short-term changes in Da response in certain cells in the reward network, and affects learning and cognitive performance, but it also causes long-term changes in the number and functioning of receptors, which in turn potentiates changes in behavioral traits (e.g., cocaine-seeking). These changes entail changes in gene expression. (Notice that I'm not even talking about epigenesis: epigenesis further strips any possible remaining shreds of meaning from the concept "innate," but we don't even need to evoke it to do away with the construct entirely. Good old gene expression will do.)

Of course solving math problems is a very different sort of trait than a limb. Still, it's a mistake to think of a limb as a fixed or predetermined environment. Certain features of limbs (e.g., number of muscle fibers) are fixed after a certain point in development (well, they can be lost, but not gained); others are dynamic (e.g., bone remodeling; hematopoietic processes). But it's hard to map a complex behavior onto a complex anatomical structure. One of the difficulties is: we might all agree that an arm is a trait of an organsim (lizards got 'em, snakes not so much). Is going to band practice after school a trait of my son? Hm...it's a behavior, but a trait? He does it regularly, it's very structured and deliberate, perhaps even adaptive somehow. What about "bipedal locomotion?" Is that the right level of generality for calling something a trait? Is bipedal locomotion innate in humans? That's where things really fall apart: all sorts of things can happen in development  -- experiences -- that alter bipedal locomotion. If one said "Well, snakes are not innately capable of bipedal locomotion, but humans are," it's sort of hard to argue with that. But you also haven't really said anything. There's no explanation for either ontogeny or phylogeny of the trait. "Innate" carries no causal, and precious little descriptive, information.

Take a harder case: a child identified as "good at math" vs. one "bad at math," from the tenderest age. Do they have innately different skills? Well, what do you mean? At some point, the differences in mathematical thinking will be embodied in brain differences; perhaps these differences had already emerged before their earliest experiences with what we might call quantification or pre-math education. If so, were these differences "because of" differences in their genotypes? In what way?

Perhaps the *only* sensible way a word like "innate" could be used is as a near-synonym for "in the genome" to refer to specific genes (or lack thereof, in deletions) or more commonly, alleles of specific genes. So it might be reasonable to use "innate" as another way of saying, e.g., "has the A allele of SNP 4 of GRM3" (a particular allele of a glutamate-receptor encoding gene." So you could say "He is innately GRM3 SNP4 A/A" or "He is GRM3 SMP4 homozygous." The latter is how geneticists would typically say it, I think. Again, the word "innate" adds no information, and introduces a bit of confusion (because it is so often used incorrectly), and there is precise disciplinary terminology available instead.

Once you deviate from that usage, you're in a minefield of poor assumptions. For example, if you say "He's innately predisposed to develop schizophrenia" because he (whoever he is) has the SMP4 A allele, then you'd be saying something misleading: GRM3 is one of several statistically significant but individually weak predictor of schizophrenia (and some cognitive phenotypes), but not everyone w/ the allele will talk to shrubs, nor is the allele an individually strong predictor of any cognitive test outcome as-yet discovered. Same for every other genetic marker found to be associated with any other common neurological or psychiatric disorder, or normally varying cognitive or behavioral phenotype, as yet investigated (to my knowledge). Of course genetic variables contribute to all of these traits, but no information is added, only confusion, by adding the word "innate" to the description of the trait or a quality of the trait. If you assume that genes are involved in anatomy and physiology -- that is, if you accept the most basic premises of 20th/21st-century biological sciences -- then it is assumed.

Hope that clears up my point.

From Simon's original question:

"This allows that representation may not be an innate capability, but a property of human culture, propagated via cultural training.What evidence do we have, ie in developmental psychology or anthropology or neurology, that internal representation is either innate or learned?"

Can we point to decisive evidence that internal representation is an innate capability of the human brain? I suggest that the SMTT experiment, illustrated below, provides this kind of critical evidence. In the SMTT experiment, a vivid hallucination of a triangle in horizontal motion is experienced by the observer when there is no such object in the visual field. There appears to be no way that such a vivid visual representation could be the result of learning. On the other hand, a detailed neuronal model of what must be an innate brain mechanism (the putative retinoid system) successfully predicts the features of the induced hallucination. Of course, this does not preclude the fact that the brain also has the capability of learning new representations. But even in this case, the capability for this kind of learning must be based on particular kinds of brain mechanisms that are innate.

Arnold, those sorts of optical illusions are sometimes given as examples of "innate" mechanisms. I would argue that although they are high-probability phenotypes of the developing visual system, it is not informative in any way to say they are "innate." They are, like all other visual phenotypes, momentary (though relatively stable) dynamic properties that are products of a history of developmental experience of the various components of the visual system, all of which gradually emerged in an EXPERIENCE-EXPECTANT manner. Here it is useful to think of "learning" as a new construct: our folk-psychological concept "learning" is prototypically associated with book studying, school, declarative knowledge, etc. However, if you replace "learning" with "experience," consider that the patterns of connectivity among bipolar, etc. cells, and the effects of output of ganglion cells to cortical columns are highly experience expectant. So the firing patterns that give rise to such illusions are emergent properties of a healthy visual system developing under a fairly typical regimen of experience, exposure to nutrients & oxygen, and a inter- and intra-cellular chemical environment that is partly determined by changes in gene expression during the early development of the system. What part of that, precisely, is "innate?" The gene expression? If so, than "innate" just means "is a property of a biological system" or something like that. In that case every trait of trees, fungi, weasels, yeast, etc., is somehow "innate," and every feature of cars, table, rocks and socks is not. Ok, we can agree on that. But then the word, and construct, carries no real information. That's ok for some words -- people use a lot of words that don't carry much specific semantic meaning -- but the problem is that people, including scientists, use the word innate AS IF it carries meaning -- as if it specifies some quality of a trait, or some causal sequence or origin. Therein lies the problem. I happen to believe that scientists, of any occupations, should be precise in how they use language in professional communication, and think through the meanings of the theoretical constructs they use.

Gedeon: "... but the problem is that people, including scientists, use the word innate AS IF it carries meaning"

The word innate does carry meaning in the context of a biological understanding of mental representation. For me, and other neuroscientists of my acquaintance, it qualifies the essential brain mechanisms that enable mental representation as being genetically endowed rather than being brain structures that are created in the neuronal processes of learning. When we talk about a property of the brain being innate or learned, we are engaged in meaningful communication. I think that however precise we wish our language to be, meaning is always dependent on the context of use and the background of the communicants. For example, see The Pragmatics of Cognition in "Overview and Reflections" on my RG page.

I should also stress that the SMTT experience is not an illusion; it is a true hallucination. Therein lies its significance for our understanding of conscious representations.

Arnold and Gedeon,

thankyou so much for your stimulating and informative contributions to this thread. Before I focus on specific points, I would like to note (for readers of the thread as well as for us) that in your exchanges, I see a paradigmatic example of a Kuhnian paradigm shift. (Though Kuhnians and anti and post Kuhnians may take me to task on this:)

Part 1 - world views in conflict

Terminology that Arnold takes as axiomatically meaningful is found scientifically empty by Gedeon. If nothing else, this should function as a parable on the importance of philosophy in science and a caution to empirical pragmatists. Gedeon, I agree that terms like 'trait' are due for a thorough spring cleaning.

It is also a fine example of the slow but shifting tides of inter and trans (etc) disciplinarity in the academy. Gedeon's terms of reference - ie genetics and molecular biology - were not so long ago regarded as so far outside the territory of cognitive science as to be irrelevant.

The biological materialism reflected in Gedeon's comments is a comparatively new force in Cog Sci. Likewise is the integration of 'brain science' with embodied and developmental observation/experiment. This maps onto, or is the human dimension of neuro-ethology, which likewise sought to avoid the excesses of internalism and the imprecision of externalist ethology, and by combining the two, have corrected many errors'. I'm reminded of a related example - Tony Chemero's analysis and critique of rat experiments from a Gibsonian Affordance viewpoint.

The situation that you identify wrt 'innate', and 'trait' (and doubtless many other terms, ie 'representation', 'symbol' etc) is, I would say, characteristic of a discipline in decline, or about to decline. Off the cuff, let me sketch an outline of the life history of a 'discipline', based in part on a quote from Richard Rorty, which I'm quite fond of -

“Interesting philosophy is rarely an examination of the pros and cons of a thesis. Usually, it is implicitly or explicitly, a contest between an entrenched vocabulary which has become a nuisance and a half-formed new vocabulary which vaguely promises great things.”

Lets say there are five stages -

"The five stages of the life history of a 'discipline',

1. the stage of half-formed vocabulary and vague promises

2. The stage in which ideas from various fields are interfaced, and new relationships and distinctions are built, out of which, assiduously and incisively, a new vocabulary, reflective of a new world view, is built.

3. The stage when this vocabulary is deployed by members of the group in order to conduct sophisticated and dense discourse(s). In this stage the epistemological history of these terms is shared knowledge and the terms operate as shorthand

4. These terms, rather than standing for a history of research and debate become reified, and, for instance, are written about in text books and taught, to a new generation of students who take these ideas as axiomatic ground-level realities. As a result, the terminology so rigoroulsy developed in 2+3 become like magical incantations full of presumed meaning.

5. The stage of paradigmatic failure, where problems arise which often appear to need, ie, only technical or methodological tweaks, but as problems progress, turn out to be problems in principle. The explanatory power of the paradigm comes into question and interrogation, internal and external begins.

return to 1 and repeat.

PART 2 - our conversation

2.1 Arnold said "the immediate global unity of the egocentric conscious presence"

To ask what this sensation is, is to go to the root of philosophy of mind. Thinkers like Dorothée Legrand, Susan Hurley, Evan Thompson et al will argue that such a sensation is bodily. In any case I see no reason to assume - contra Andy Clark and Jonathan Edwards - that the sense of 'egocentric conscious presence' is located in the brain, let alone at a tiny point (pineal gland?) or exclusively in the brain. Depending on your definition of cognitive, that would make it either a matter of cognition or not. I tend to the 'not' side.

2.2 in the same passage, Arnold said "Grid cells and place cells are part of the machinery of spatial cognition but they do not represent our enveloping phenomenal world"

This implies that you subscribe to a somewhat mechanistic idea that component operations occur somewhere, and are integrated somewhere else. I think we have to be very careful of the metaphors we deploy, and especially the very common industrial subassembly metaphor. Gallese and Lakoff have argued - fairly persuasively in my opinion - that this is not the case, at least in terms of the transition from sensorimotor routines to (linguistic) concepts.

2.3. Arnold, in his next message, said - "Mental representation is not just a trait. Mental representations are biophysical events produced by particular kinds of innate brain mechanisms."

This is, again, the question I am asking. Arnold - for me, your response asks two fundamental questions. the first is, neurologically, what and where are those biophysical event? Has anyone come up with a neural phenomenon which is generally descriptive/explanatory of mental representation at the biological level?

Second - are those mechanisms "innate" in the sense it seems you are asserting, ie are they part of the basic biological equipment of the species, like thumbs or eyes? I am inclined to think not, based on ideas of neural exploitation (Bates, Gallese, etc). And as noted in my original question, Malafouris has made what I think is a groundbreaking stab at an enactive, embodied explanation of the possible origins of mental representation. Of course, he is using representation (only) in the sense of pictures - which identifies his argument with conventional visuo-centric approaches. I am looking for a more encompassing conception of representation, including multimodal and sensorimotor representation.

2.3 Gedeon: "all traits, including anatomical traits, are emergent products of a history of exposure to environments that change the physiology and structure of the cells involved."

So all 'traits' are reducible to events at a cellular, or genetic/epigenetic level and none are 'emergent'?

2.4 " Is going to band practice after school a trait of my son? Hm...it's a behavior, but a trait? He does it regularly, it's very structured and deliberate, perhaps even adaptive somehow. "

so aren't you already deconstructing 'trait'?

2.5 "What about "bipedal locomotion?" Is that the right level of generality for calling something a trait? Is bipedal locomotion innate in humans? That's where things really fall apart: all sorts of things can happen in development -- experiences -- that alter bipedal locomotion."

Right,and as Mauss tells us, walking is cultural. Different cultures walk in different ways, sit in different ways, etc, which brings us to the matter of culturally determined affordances. A chair affords sitting to us but not to a traditional japanese. An east african carved wooden neck-rest does not afford comfortable rest to us in the way a pillow does.

2.6 "At some point, the differences in mathematical thinking will be embodied in brain differences"

but beware of attribution error! To say certain neurological qualities concur with mathematical ability is not to say that mathematical ability is (wholly) in those neurological qualities.

2.7 "Perhaps the *only* sensible way a word like "innate" could be used is as a near-synonym for "in the genome""

This makes you a genetic essentialist?

2.8 Arnold "Can we point to decisive evidence that internal representation is an innate capability of the human brain?"

Like Gedeon, I am dubious of the inferences you make based on SMTT.

Also, I think there are other question that must be asked in order that

the inferences from SMTT stand. These would include -

a. "Can we point to decisive evidence that internal representation is NOT, or is NOT ENTIRELY an innate capability of the human brain?"

b. "Can we point to evidence that activities which were previously assumed to require internal representation in fact do not?"

I think the answer to both these is likely to be affirmative.

2.9 Gedeon - "the firing patterns that give rise to such illusions are emergent properties of a healthy visual system developing under a fairly typical regimen of experience"

Right, we might say experience-expectant or we might say predictive/anticipatory. In evolutionary terms, its clear why we have these mechanisms. I hate to resort to the hoary old sabre tooth tiger and caveman story, but, a snapping twig behind you at night might just be a snapping twig, or it might mean imminent pain or death.

2.10 Arnold - "the essential brain mechanisms that enable mental representation as being genetically endowed rather than being brain structures that are created in the neuronal processes of learning."

Thankyou for bringing us back to the question :)

2.11 Arnold "I should also stress that the SMTT experience is not an illusion; it is a true hallucination."

Your insistence on a clear distinction between the two is, if you permit me, a case of point 3 in my The five stages of the life history of a 'discipline'. But that distinction is contested - ie my point 5.

In the SMTT case, there is some 'real world' stimulus. The idea of the slot and partially occluded object is not unlike the hand grasping the beer bottle 'illusion' that Alva Noe refers to (I can't bring to mind the originator) - eyes closed, grasp a beer bottle. We assume it is a bottle, and not just a cutout of a cylinder with surface only under our fingers. We anticipate or predict a bottle. How is the case of your triangle different?

PART 3 - The question of mental representation

Arnold believes in something he calls mental representation, and believes it is genetically given rather than cultural. I would counter that the very fact that different human cultures have different systems of (external) pictorial representation. Some cultures seem not to have any. We are highly acculturated to photographic representation which is by no means a natural or normal kind of representation. This would seem to suggest that human cultures develop mental pictorial representation in different kinds and amounts. The (A) question then would be what is the neurological foundation which permits these different kinds of representational systems?

Gedeon also refers to mental representation and seems to affirm their existence. But I remain unclear about the phenomenon we are discussing.

When I close my eyes, I can see the face of my sweetheart in my 'mind's eye'. I could even sketch it 'from memory'. That, in an experiential sense, is all the proof I need of the existence of 'visual' mental representation,ie pictorial memory. But how is it neurologically constituted? How do we come to have the ability ? - to varying degrees, and - can this notion of representation be generalised to other modalities and combinations of modalities.

This is beginning to sound like the debates over 'representation' which contributed to the collapse of "good old fashioned AI" in the late 80s. If so, it seems like it is a question which as yet has not been satisfactorily addressed.

best,

Simon

Simon,

Again, thanks for your detailed comments. Foundational disputes, such as appear on this thread, are to be expected on the frontiers of science where, surely, mental representation is located. Your post raised many interesting questions. In this reply, I'll touch on just a few.

Simon: "In any case I see no reason to assume - contra Andy Clark and Jonathan Edwards - that the sense of 'egocentric conscious presence' is located in the brain"

There is a vast amount of empirical evidence that consciousness of any kind is located in the neuronal mechanisms of the  brain. What evidence can you cite that locates consciousness in any other part of the body?

You wrote: "This implies that you subscribe to a somewhat mechanistic idea that component operations occur somewhere, and are integrated somewhere else. I think we have to be very careful of the metaphors we deploy ..... Gallese and Lakoff have argued - fairly persuasively in my opinion - that this is not the case, at least in terms of the transition from sensorimotor routines to (linguistic) concepts."

I had a discussion with Lakoff on EDGE that relates to this point. Here is what he said in response to my comments:

"These are cases where it appears that the structure of the brain imposes "sharp limits" on conceptual structure.

These results are very much in the spirit of Professor Trehub's very interesting observations and his claim that "detailed minimal models of cognitively relevant neuronal mechanisms are required if we are to understand the constraints operating on the brain/mind." That is just what we are finding from the computational neural modeling perspective."

So it appears that mechanistic explanations (neuronal mechanisms) of linguistic concepts are consistent with his own findings. The claim that sensorimotor routines, which depend on whole-body mechanisms that input to the brain mechanisms that integrate these afferent signals, will give us many of our linguistic concepts, seems well founded.

Simon: "In the SMTT case, there is some 'real world' stimulus. The idea of the slot and partially occluded object is not unlike the hand grasping the beer bottle 'illusion' that Alva Noe refers to (I can't bring to mind the originator) - eyes closed, grasp a beer bottle. We assume it is a bottle, and not just a cutout of a cylinder with surface only under our fingers. We anticipate or predict a bottle. How is the case of your triangle different?" 

It is really important to appreciate the difference between what we anticipate, predict, or imagine, and what we actually SEE. In the case of the SMTT hallucination, the subject does not anticipate or imagine a triangle. A vertically oscillating dot is seen by the observer, then it suddenly disappears and is replaced by a vivid conscious experience of a real triangle oscillating horizontally. The triangle appears to be really out there in front of the observer, even though there is no such retinal image. In fact, by turning a knob,  the observer is able to control the width of the hallucinated triangle to match its height when the height of the unseen vertically oscillating dot changes. The successful match of the width and height of the hallucinated triangle is verified by independent observers who experience the same hallucination when they look over the shoulder of the subject.

Simon: "Arnold believes in something he calls mental representation, and believes it is genetically given rather than cultural. I would counter that the very fact that different human cultures have different systems of (external) pictorial representation. Some cultures seem not to have any. ....... The (A) question then would be what is the neurological foundation which permits these different kinds of representational systems?"

We shouldn't conflate the question of the biological existence of mental representations with questions relating to the format or content of mental representations. As for their neurological foundations, I suggest that you look at "Learning, Imagery, Tokens and Types: The Synaptic Matrix" and "Modeling the World, Locating the Self, and Selective Attention: The Retinoid System" on my RG page.

I agree with Professor Arnold Trehub that we must exercise critical scientific framework when working through a biological paradigm for the mental representations  against the concept of  neuronal processes with regard to mental representations  

@  Simon Penny and Arnold Trehub

Prof Trehub writes: Representations are distinct patterns of neuronal excitation in the brain that are systematically evoked by real objects and events, or by imagined objects and events.

I think the issue of internal representations needs to be better defined before we can progress with the discussion. As usual, different participant talk about representations of different scales. As a result they cannot agree on anything.

I would start with the most fundamental scale of representation at the level of individual sensory cells rather than their integrations. At this level all representations in the cortical maps replicate only the relationship within the spatially distributed (let us call it) “something out there”. As illustrated in the figure below, what is duplicated is not the objects itself but the spatial relationships within the 'object' the body interacts with.

                                                  ∎               ☻

                                                  □                ☺

      SOMETHING OUT THERE □                ☺ SENSORY MAPS

                                                  ∎                ☻

                                                  □                ☺

These spatial relations are recorded only in terms of the sensory endings capable of interacting with the object's properties. This means that we know only a few selected aspects of “something out there” and only in terms of the spatial effects this “something out there” has on the sensory endings we have. Note that one can duplicate spatial relations with everything – e.g. with Lego, sensors in the digital camera or pencil and, of course, with the activated/non-activated sensory cells represented in the brain. What matters to the being is not what is “out there” but how this something relates spatially to the being.

On this view, there are as many senses of space as there are sensory modalities and their combinations/associations. The 2-dimensional patterns are associated first within the same modality and then with other modalities to form a great variety of larger-scale notions including the planar grid representations you mentioned and the notion of "chair" or any other object for that matter.

What I propose is the first stage of space processing. We routinely talk about space and time in terms of real objects, or at best in terms of high level conceptualisations and their expressions. It is unfashionable to ‘reduce’ language concepts to sensations without being immediately accused of reductionism and god knows what else.

The views I expressed here are based on my Ph.D thesis. 

Great question.

Let me start with a small riddle... Please fill in the blanks in this text copied from a peer-reviewed article published in 2002:

The inspired guess is that many people, irrespective of their background, would fill in brain. This article must have been about the brain-as-a-computer-metaphor others have referred to and the author apparently does not want to compare the brain to what goes on inside a human built desktop computer except... it had to be universe.

The text is from an article entitled Computational Capacity of the Universe by Lloyd (2002). As is evident from the quoted text, concepts such as computation, information processing and information representation are not supposed to be interpreted in terms of what goes on in a digital desktop computer when describing the behaviour of a natural physical system.

Physical information (at the meso scale) is a measurable quantity that resolves uncertainty about a system state. Information representation in physical systems refers generally to the entropy associated with a global system state: The information that would be needed to provide a complete description of all the different micro state configurations that can generate the macro state of the system.  Information processing refers to order generation, that is, given a system that transitions from one state into another, the information needed to describe the new state of the system will have changed. This change in information as a measure of uncertainty is called processing or computation.

What has happened is that behavioural and cognitive sciences have adopted an unauthorised version of information theory (Shannon, 1948), one in which information representation is conflated with meaning and computation is considered an act of coding and decoding by a logical structure. If the social sciences indeed suffer from a physics envy, they most certainly have a blinding information infatuation, because meaningful information, or message content is NOT a topic of information theory:

The original question was wether internal representations could be innate or learned. The explanatory work such a proposition does is to describe how the contents of a message can persist in time. As Shannon indicates: The reproduction of messages is a fundamental topic of information theory, but this cannot be based on specific message content, because at the time of design it is unknown which messages have to be reproduced.

Of course, at some point one wants to make reference to meaningful information when discussing the behaviour of living systems. Perhaps the safest bet is to suggest that what is reproduced, is a "design", a structure that permits the selection of messages based on their ecological relevance, which is in fact a definition of intelligence (inter-ligere: linking things together to make sense of the message content; see Desurvire, 2009, pp.38). The purpose of intelligence in all the different contexts the word can be used in, from secret service agencies to scientists studying the behaviour of living systems, is: Control the flow of information change in a dynamical physical environment in order to select what is relevant given current needs, expectations and priorities. 

The ecological approach to explaining complex adaptive behaviour has developed many explanatory tools that can deal with such things (see Turvey & Carello, 2012). They may also have developed an information intolerance over the years, so some of the theoretical constructs may take some getting used to. 

All the best,

Fred

Desurvire, E. (2009). Classical and Quantum Information Theory: An Introduction for the Telecom Scientist. Cambridge University Press.

Lloyd, S. (2002). Computational Capacity of the Universe. Physical Review Letters, 88(23), 237901. doi:10.1103/PhysRevLett.88.237901

Shannon, C. E. (1948). A Mathematical Theory of Communication. Bell System Technical Journal, 27(3), 379–423. doi:10.1002/j.1538-7305.1948.tb01338.x

Turvey, M. T., & Carello, C. (2012). On Intelligence From First Principles: Guidelines for Inquiry Into the Hypothesis of Physical Intelligence (PI). Ecological Psychology, 24(1), 3–32. doi:10.1080/10407413.2012.645757

@ Fred Hasselman

Not only a great question, but also great answer. It is the first time I came across the answer which makes  sense to me. Thanks. 

Below is a short extract from my 2011 Ph.D dissertation in which I try to clarify the notion of information as that of a human internal form: 

I hope this is not too cryptic for the readers of this blog. 

Best, Wes

Fred Hasselmann -

thankyou for reminding us of the theoretical sleight of hand which one sees so often wrt Shannon's work:

"What has happened is that behavioural and cognitive sciences have adopted an unauthorised version of information theory (Shannon, 1948), one in which information representation is conflated with meaning and computation is considered an act of coding and decoding by a logical structure."

Exactly, Shannon stated explicitly that he was not making any claims wrt meaning.

thankyou

Simon

Simon,

Here's how I think about the (non-Shannonian) concept of information:

Information is any property of any object, event, or situation that can be detected, classified, measured, or described in any way.

1. The existence of information implies the existence of a complex physical system consisting of (a) a source with some kind of structured content (S), (b) a mechanism that systematically encodes the structure of S, (c) a channel that selectively directs the encoding of S, and (d) a mechanism that selectively receives and decodes the encoding of S.

2. A distinction should be drawn between latent information and what might be called kinetic information. All structured physical objects contain latent information. This is as true for undetected distant galaxies as it is for the magnetic pattern on a hard disc or the ink marks on the page of a book. Without an effective encoder, channel, and decoder, latent information never becomes kinetic information. Kinetic information is important because it enables systematic responses with respect to the source (S) or to what S signifies. None of this implies consciousness

.

3. A distinction should be drawn between kinetic information and manifest information. Manifest information is what is contained in our phenomenal experience. It is conceivable that some state-of-the-art photo—>digital translation system could output equivalent kinetic information on reading English and Russian versions of *War and Peace*, but a Russian printing of the book provides me no manifest information about the story, while an English version of the book allows me to experience the story. The “explanatory gap” is in the causal connection between kinetic information and manifest information.

In the retinoid theory of consciousness, the perspectival/subjective structure and dynamics of our brain’s representations in retinoid space provides the causal connection between kinetic and manifest information. A very large body of empirical evidence supports the retinoid theory of consciousness.

-----------

@Wes Raikowski - Is the text of your dissertation available somewhere? reference to spatial structure is very sensible and reminds me of Boyle's work on message content:

-----------

@Arnold Trehub - I personally would prefer a science of life (including mental life) that uses formal theoretical constructs that can be evaluated for their similitude to alternative claims competing to explain the same phenomena. Only then can we appraise their respective verisimilitude in a scientific sense.

If theoretical constructs cannot produce divergent predictions that suffice to announce a 'winner', they must be considered the same for all intents and purposes (e.g., wave or particle based descriptions of matter and energy). Whenever possible, such findings of similitude should lead to unification of ontology... which generally heralds the advancement of science (e.g., the quantum formalism allowed treatment of many different "pictures" of reality as different sides of the same structural coin).

I believe it is fair to say that similitude of constructs of information theory such as  information and entropy has been established, it may happen soon for free-energy / negentropy as well (see, e.g. Friston, 2012; 2013; Mahulikar & Herwig, 2009). Physical information theorists “[…] regard the physical world as made of information, with energy and matter as incidentals” (Bekenstein, 2003, p.59). This is valid in many cases, because of the formal connections that have been established.

I believe one should be extremely modest when suggesting a redefinition of highly corroborated constructs (information, force, kinetic). This is warranted only (in my opinion) if it would have the prospect of unifying different ontological frameworks. This cannot be established without practicing at least some of the formal rigor common to the natural sciences. That being said, I do not think that your conception of information contradicts, or is dissimilar from (physical) information theory or systems biology:

Information is any property of any object, event, or situation that can be detected, classified, measured, or described in any way.

Property attribution to theoretical objects of measurement by means of observation is an act of resolving uncertainty about the object, by means of assigning an amount of information to its observed state. Whether you use kilograms or words, you use an amount of quantifiable information. So in my opinion this definition is not "Non-Shannonian" at all (noting that a 1-to-1 correspondence between measurement and objects property is only unproblematic in classical physical measurements).

If you flip a fair coin, the observation of one of the two states represents maximal information (entropy) about the system, because the sates are equiprobable. Any deviation from equiprobability would be a deviation from maximal information representation. If only 1 side came up every time it was flipped, the system would represent no information at all.

"2. A distinction should be drawn between latent information and what might be called kinetic information. All structured physical objects contain latent information. This is as true for undetected distant galaxies as it is for the magnetic pattern on a hard disc or the ink marks on the page of a book. Without an effective encoder, channel, and decoder, latent information never becomes kinetic information. Kinetic information is important because it enables systematic responses with respect to the source (S) or to what S signifies. None of this implies consciousness"

I think kinetic information, in terms of the explanatory work it does here, refers to the possibility to express an analogy between different (latent) information sources. I do not think 'latent' is necessary: An information source or structure is a physical system whose complete description in terms of amount of information is subject to change. This means it can represent different amounts of information, by means of different configurations of its micro-structure (entropic states). This is the same property as indicated by 'latent': A capacity to represent / register information.

The role of the 'kinetic' information is similar to the definition of a code. In organic chemistry it is known that two (latent) information sources (nucleotides and amino acids) can be connected by a coding structure (transfer RNA) by means of the chemical operation of translation that consists of two physical recognition processes. Organic codes can give meaning to information structures by translating their similarities, that is, by recognizing the invariant structure that exists between them. 

The transfer RNA molecule connects the world of nucleotides (DNA) to the world of amino acids (Barbieri, 2003, p. 98): A message from the world of nucleotides is transmitted by the tRNA which also acts as the channel as well as the receiver (decoder) in the destination world of amino acids. 

The use of 'kinetic' to refer to the configuration of the coding structure is perhaps not the best option in this context because:

Codes give meaning to information sources that do not appear to share any identity, if no reference were made to the code itself. If a force would be responsible for the correlation to be captured by the analogy, there would be no need to map the correspondence between the structures. The code to establish a translation in that case, would be a law of physics. This defines organic memory systems capable of representing meaningful information a very special natural phenomenon, capable of accumulating accidental event occurrences, random events by turning random order into (thermodynamically) improbable order.

Establishing identity across spatial or temporal scales is an act of order generation, by means of specification of the configurations of a physical system to its history of interactions, therefore indeed: "The existence of [meaningful] information implies the existence of a complex physical system". It also does not imply any consciousness, although I would say intelligence does play a role, because the content of the message is used to 'select' it. To be clear, I consider it a self-organized intelligence, generating an improbable order of the universe.

Manifest information resembles the affordance-effectivity relation in ecological bio-physics and is in principle not any different from a code. It is a convenient way of saying: Some structural invariants in the environment provide an organism opportunities for action, for maintaining and increasing the complexity of its internal structure, but can only be defined as such when the ecological context of the animal is considered. The structures in the environment that are "climb-on-able"  or "pass-through-able" for a mouse, do not refer to the same structures to which I would attribute those properties.

It is important to keep distinguishing meaning from information, because information is a conserved quantity of the universe, it cannot be destroyed, not even by a black-hole (is what Stephen Hawking lost a bet about, Hakwing, 2005). Meaning, however, can be lost or erased and this can be achieved by using free-energy to do work and this increases the entropy of the universe (Landauer's principle, cf. Bérut et al., 2012). This points to a strong link between meaning and ordered physical structure of high thermodynamic improbability.

But neuronal reductionist models of consciousness have too many unresolved anomalies for my tastes see e.g. Majorek (2012) discussing cases such as this one: http://www.newscientist.com/article/dn12301-man-with-tiny-brain-shocks-doctors.htm

I believe we first need to establish a cross-disciplinary consensus formalism on physical information and the behaviour of complex dynamical systems in which mental phenomena may be observed before we can even begin talking about physically, biologically and psychologically plausible model of consciousness. I do not see how we can appraise competing models given current knowledge. 

Of course will gladly be shown wrong about my assertions so far.

-----------

@Simon Penny - Sorry, I did't mean to take over the thread and side-track the discussion, I hope you find the references listed here interesting in the context of your original question as well. As a good place to start for arguments against neuronal reductionism I recommend: Majorek (2012). Does the Brain Cause Conscious experience? http://mind.oxfordjournals.org/content/102/406/263.full.pdf

-----------

kind regards,

Fred

Barbieri, M. (2003). The Organic Codes: An Introduction to Semantic Biology. Cambridge University Press.

Bekenstein, J. D. (2003). Information in the Holographic Universe. Scientific American, 289(2), 58–65. doi:10.1038/scientificamerican0803-58

Bérut, A., Arakelyan, A., Petrosyan, A., Ciliberto, S., Dillenschneider, R., & Lutz, E. (2012). Experimental verification of Landauer’s principle linking information and thermodynamics. Nature, 483(7388), 187–9. doi:10.1038/nature10872

Boyle, C. F. (1992). Physical Laws And Information Content. In Workshop on Physics and Computation (pp. 52–57). IEEE. doi:10.1109/PHYCMP.1992.615491

Friston, K. (2012). A Free Energy Principle for Biological Systems. Entropy, 14(11), 2100–2121. doi:10.3390/e14112100

Friston, K. (2013). Life as we know it. Journal of the Royal Society, Interface / the Royal Society, 10(86), 20130475. doi:10.1098/rsif.2013.0475

Hawking, S. (2005). Information loss in black holes. Physical Review D, 72(8), 084013. doi:10.1103/PhysRevD.72.084013

Mahulikar, S. P., & Herwig, H. (2009). Exact thermodynamic principles for dynamic order existence and evolution in chaos. Chaos, Solitons & Fractals, 41(4), 1939–1948. doi:10.1016/j.chaos.2008.07.051

Majorek, M. (2012). Does the Brain Cause Conscious experience? Journal of Consciousness Studies, (3), 121–144. Retrieved from http://mind.oxfordjournals.org/content/102/406/263.full.pdf

Walker, I. (1983). Complex-irreversibility and evolution. Experientia, 39, 806–813. Retrieved from http://link.springer.com/article/10.1007/BF01990396

Fred: "I believe we first need to establish a cross-disciplinary consensus formalism on physical information and the behaviour of complex dynamical systems in which mental phenomena may be observed before we can even begin talking about physically, biologically and psychologically plausible model of consciousness. I do not see how we can appraise competing models given current knowledge."

I disagree. We each have our personal intuitions but, in science, evidence trumps intuition. You might take a look at "Evolution's Gift: Subjectivity and the Phenomenal World" and "A Foundation for the Scientific Study of Consciousness" on my RG page. I think we can appraise competing models of consciousness within the norms of science. For example, what model, other than the retinoid model, can predict the vivid visual hallucinations and their controlled variations demonstrated in the SMTT experiments?

@Arnold Trehub - I am familiar with the retinoid model described in Trehub (2007). In the absence of any formalism, the appraisal question should be: "What would be the evidence that would convinve you the retinoid model is false?"

Trehub (2007) mentions Einsteins classical tests of General relativity: "A famous example is the observation, in 1919, of a displacement of the light from Mercury as it grazed the eclipsed sun" this must refer to the 1919 Eddington expedition to confirm the bending of starlight due to the gravitational field of the sun by simultaneous observations in Africa and Brazil.

Mercury had absolutely nothing to do with that test of GRT. The accurate 'prediction' of Mercury's perihelion advance, an anomaly to Newton's theory, was a very convincing corroboration of GRT (see Hasselman, 2013; http://journal.frontiersin.org/Journal/10.3389/fphys.2013.00075/full ).

The three classical tests Einstein proposed in 1916 can be found here: http://en.m.wikipedia.org/wiki/Tests_of_general_relativity#Classical_tests

Einstein wrote about any failing predictions:

"The chief attraction of the theory lies in its logical completeness. If a single one of the conclusions drawn from it proves wrong, it must be given up; to modify it without destroying the whole structure seems to be impossible"

So, after the announcement of a number of crucial predictions that have to be passed or otherwise falsify the claims, one would want to see a track-record of succesfully passed rigorous tests of thise predictions.

The SMTT experiments and moon illusion do not comprise a convincing track-record of the retinoid model to me, even though it may be biologically plausible.  They certainly  do not compare to the predictions by GRT. To be clear, these predictions concerned measurement outcomes. What were the predicted measurement outcomes that estsblished the empirical accuracy of the retinoid model in the SMTT experiments?

Perhaps I am too critical in these matters, but I believe anything can be modelled ex post-facto and is as such unremarkable. What behavioral cognitive science needs, is a program of strong inference in which divergent predictions are rigorously tested (see Hasselman, 2013).

I will have look at the other literature you suggested,

kind regards,

Fred

@ Fred Hasselman,

Thank you for the ideas which I find very interesting. Here is my reply to your question: 

Regarding: “Reference to spatial structure is very sensible and reminds me of Boyle's work on message content:”

My dissertation is a philosophical take on cognition understood as a physical system which evolved to monitor movements of the body within its external environment. It postulates the existence of conceptual understracture common to all human and non-human expressions.

My approach, unlike that of Professor Arnold Trehub, is a low level investigation in the sense of the sensory organisation before it is integrated in higher cognitive areas. I am basically interested in the first stages of the interaction between, what I call, “something out there” and the endings of sensory cells embedded within the body or, in other words, in how (and why) the activation of those cells is organised in the brain in the form of two-dimensional maps.

Instead of neuroimaging, I looked for the evidence in human expressions found in language, artefacts, mathematics, biology and physics, etc. The analysis of those expressions suggests that sensory cells are organised into cortical and other maps to code the spatial relations (which are essential for movement) within the body. The patterns in these maps are first associated within sensory modalities and then between different modalities to be finally associated with the motor maps.

I have very little to say about higher order associations such as the subject of consciousness discussed by Arnold, but I believe that there is no contradiction between his account and mine as they are two views of the same system seen from different scales. You will be pleased to know that my Ph.D thesis was greatly influenced by “Evolving Hierachical Systems” written be Stanley N Salthe, a great biologist, before he turned his interests to semiotic theories of biology, which I think was a mistake.

During my research I noticed that symbols used in human expressions are arbitrary (different languages use different sounds and marks) but the ways those symbols are combined are very stable across time and different cultures. These combinations, which are called concepts, make possible for us to understand each other, translate between languages and understand the languages which do not longer exist.

For example, the concept “information” as used across ages retains the core notion of “form” and prefix “in” which suggests the internal aspect. The word form is an intuitive concept based on the sensation of individual’s own body. It involves the content (e.g. cells) and space. Form without space is understood as content. An external view of one’s own form is called “figure”. And the view of other figures/shapes is perceived as a context of one's own figure.

All these terms originate from the way our senses are organised – interoceptive senses give us an access to our internal form, and vision provides us with the sense of own figure and its context. The sense of touch is unique as it provides the access to both the sense of own figure (we can explore own body by touching) and own form (sense of being touched). As such it could be used to substitute for vision. And so on.

The notion of internal form is interesting because it is not accessible to direct measurement as it comes as a single integrated whole. In addition to content, it involves space which allows the content to be arranged in different ways we call forms. Each form has its own set of spatial/relational experiences which we tend to conceptualise as “information”. I read somewhere that Shannon intended to call his definition of information as “entropy”, the word which better reflects the nature of the concept he introduced.

Regarding: “Is the text of your dissertation available somewhere?”

Yes. It is available from Amazon under the title “Conceptual Understructure of Human Experience” written under the alias Wes Raykowski to make the pronunciation of my true name a little bit easier for English speakers. But before you rush to buy the book check the following site:

           https://www.youtube.com/channel/UCyzQtjyWb9itR4dGykpWwKw    

with video animations which I produced to aid understanding of some aspects of my approach.

My best regards, Wes

Fred: "In the absence of any formalism, the appraisal question should be: 'What would be the evidence that would convinve [convince] you the retinoid model is false?'"

I don't think that science can say what is true and what is false. What is true and false can only be determined in formal systems. Science is a pragmatic enterprise where both theory and evidence are subject to some degree of error. So, to answer your question, I will refer to the SMTT experiments and give some examples of possible results that would be evidence against the retinoid model of consciousness:

1. If independent observers, looking over the subject's shoulder, did not also hallucinate a complete triangle and its systematic transformations as reported by the subject.

2. If the subject were unable to match the base of the hallucinated triangle with its height.

In addition, if the predicted pendulum illusion failed, I would take this as strong evidence against the retinoid model. See "The Pendulum Illusion ..." on my RG page.

You wrote: "The SMTT experiments and moon illusion do not comprise a convincing track-record of the retinoid model to me, even though it may be biologically plausible. They certainly do not compare to the predictions by GRT."

I don't think that you can expect that a biologically-based model of first-person experience/consciousness can have the quantitative precision of a theoretical model in physics. A model of consciousness has to explain the existence of qualitative features by means of a bridging principle that relates these features to quantitative measures. The science of consciousness is in its early days, while the the science of physics has a long history. But in both scientific endeavors empirical evidence trumps intuition.

Fred: "What were the predicted measurement outcomes that established the empirical accuracy of the retinoid model in the SMTT experiments?"

1. The base width of the hallucinated horizontally oscillating triangle was a negative monotonic function of the frequency of the vertically oscillating dot. Confirmed.

2. At any given rate of the dot's vertical oscillation the base width of the hallucinated triangle was a positive monotonic function of the width of the slit. Confirmed.

3. When a figure was presented behind a triangular shaped aperture it was perceived to be swinging like a pendulum with its horizontal traverse along its vertical axis a positive monotonic function of the width of the aperture from its vertex to its base. Confirmed.

@ Arnold Trehub:

Science may be a pragmatic enterprise, but the pragmatism lies in finding a method to decide which out of several potentially true formal accounts that lay explanatory claim to a structure in reality, has highest truth-likeness, verisimilitude (or: strong inference, Platt, 1964). Why argue that the explanations of the retinoid model have "truth" to them? Because deciding on verisimilitude is the most fundamental drive behind  advancing scientific knowledge about the world.

The scientific method cannot function properly if one is not committed to falsifying one's own formal claims (Feynman: "prove yourself wrong"). At the same time, one can acknowledge that this ontological commitment is just a temporary tunnel vision. This is the position in philosophy of science known as structural realism ( see http://plato.stanford.edu/entries/structural-realism/ ). It is nicely exemplified in this quote by Poincaré om the existence of the  Æther:

Indeed, the Æther was thrown aside as useless, because the empirical studies devised to test predictions by the Æther-based models yielded meticulous null results (Michelson, 1881; Michelson & Morley, 1887). In other cases, parts of a formal system of propositions may be retained, or reframed as a special instance of a more general set of proposition. The Structural Realist's take home message is: 

It is interesting Poincaré makes an exception for the existence of material objects this is a reflection of James' Psychologist Fallacy and has some consequences for theorizing about conscious experience I call Blinded by Downward Seepage see Hasselman, 2014 ( http://dx.doi.org/10.6084/m9.figshare.1066522 )

The very first sentences of the very first chapter, of the one of the first principles of modern psychology already provide a complete description of this problem:

To be fair, the requirement to compare to physical theories such as GRT was made in Trehub (2007), I mentioned the corroborated prediction by GRT was incorrectly described in the paper. That being said, I do expect that a biological model is ultimately physical and must obey physical principle, or explain in a convincing way, why it does not..

As for the "juvenile" argument for the science that studies mental phenomena, see the attached figure (I forgot to include Herbart, 1824). Modern "exact" physics began with Maxwell's (1865) unification of electricity and magnetism and physics only became "insanely exact" after the the Quantum Formalism was established (see attachment: "the great schism" of 1925-1935")

Illusion can be helpful to understand properties of the visual system, but not (mediative) mental representations:

What appears to be a central position in the retinoid model is the detection of invariant structure between physical changes in the environment and properties of the experienced environment. This is basically what has been studied in ecological psychology for many decades, for instance: The judgement about the ability to climb a staircase is an invariant relation between the height of the riser and the length of a participants' leg (Warren, 1984, http://www.cog.brown.edu/research/ven_lab/Publications/Warren_Stairs_JEP84.pdf ) The individual shape and form of the body and the degrees of freedom available for generating behavior are what determines perception and action and ultimately the contents of the mind (also see motor imagery used as an empirical  paradigm for embodied awareness: http://psycontent.metapress.com/content/f715lx060112345k/?genre=article&id=doi%3a10.1027%2f1618-3169%2fa000237 )

More interesting when considering retinal images: The "outfielder problem" or the question that arises when trying to understand how one is able to run towards, and catch a fly-ball (e.g., McBeath et al., 1995). It turns out that the best strategy is to move into a direction that keeps the retinal image constant, thus the trajectory of the outfielder is guided by a perceptual invariant, the constancy of relative angle of motion (also why they tend to keep looking at the ball while running). No need to internalize anything and presume computations of trajectories using newtonian mechanics.

The gross "form" of our bodies may be innate, but the specifics of that form develop in a unique way (even in"identical" twins, e.g.http://onlinelibrary.wiley.com/doi/10.1002/ajmg.c.30206/abstract  ).

In a very direct way, those differences represent a physical trace of a history of interaction events that specified the structure of the organism. What is permitted in terms of variations to the gross "form" is probably innate, but that is about as much as one can say about innate representation of form and function playing a role in determining second-to-second experience and coordination of behavior.

best,

Fred

Hasselman, F. (2014): Sane As It Ever Was: The Historical Meaning of the Crisis in Psychology. figshare. http://dx.doi.org/10.6084/m9.figshare.1066522

Herbart, J.F. (1824) Psychologie als Wissenschaft. Teil 1 & 2. Königsberg. Retrieved from: http://www.deutschestextarchiv.de/book/show/herbart_psychologie01_1824

McBeath, M. K., Shaffer, D. M., & Kaiser, M. K. (1995). How baseball outfielders determine where to run to catch fly balls. SCIENCE-NEW YORK THEN WASHINGTON-, 569-569.

Michelson, A. . (1881). The Relative Motion of the Earth and the Luminiferous Ether. American Journal of Science, 22(128), 120–129. Retrieved from http://www.archive.org/details/americanjournal62unkngoog

Michelson, A. ., & Morley, E. W. (1887). On the Relative Motion of the Earth and the Luminiferous Ether. American Journal of Science, 34(203), 333–345. Retrieved from http://www.aip.org/history/gap/PDF/michelson.pdf

Platt, J. (1964). Strong Inference. Science, 146(3642), 347–353. Retrieved from http://clustertwo.org/articles/Strong Inference (Platt).pdf

Poincaré, H. (1905). Science and Hypothesis. New York: The Walter Scott Publishing Co., LTD. Retrieved from http://www.archive.org/details/scienceandhypoth00poinuoft

Fred: "Science may be a pragmatic enterprise, but the pragmatism lies in finding a method to decide which out of several potentially true formal accounts that lay explanatory claim to a structure in reality, has highest truth-likeness, verisimilitude (or: strong inference, Platt, 1964)."

So I take it that you agree that science does not announce the TRUTH, but rather what seems to be the best inference given the evidence.

Fred: "To be fair, the requirement to compare to physical theories such as GRT was made in Trehub (2007) ..."

The point made in Trehub (2007) was not to compare the quantitative precision of biologically based models of consciousness, but to compare the power of counter-intuitive theoretical prediction as in the case of GRT and, I suggest, the moon illusion and the SMTT hallucination.

Fred: "... illusions create true visual experiences under artificially constrained circumstances and are only problems for a perceiver when the constraint is successfully concealed."

I confess that I don't understand this statement. For example, the moon illusion requires no artificial constraint to be experienced. And even when the facts related to the illusion are revealed/explained to the observer the illusion persists. The SMTT experience is not an illusion; it is a vivid hallucination.

Fred: "What appears to be a central position in the retinoid model is the detection of invariant structure between physical changes in the environment and properties of the experienced environment."

The moon illusion and the SMTT hallucination demonstrate that the neuronal structure and dynamics of particular kinds of brain mechanisms systematically determine the properties of the consciously experienced environment.

Fred: "No need [for an outfielder] to internalize anything and presume computations of trajectories using newtonian mechanics."

This simplistic cartoon ignores too much! The outfielder's brain does not need to perform newtonian computations, but it does need to contain a spatio-temporal representation of the location of the outfielder in relation to home plate in order to throw out the runner on third base as he tries to run home after the catch. 

Fred: "What is permitted in terms of variations to the gross "form" is probably innate, but that is about as much as one can say about innate representation of form and function playing a role in determining second-to-second experience and coordination of behavior."

When you use the word "form", do you mean that the neuronal details of essential brain mechanisms are included as "forms"? If not, what role do brain mechanisms serve in your scheme?

Hi Simon,

Did not have time to fully read this thread but your question made me think of two interesting works.

One is a famous discussion between Fodor and Piaget edited into a book by Piatelli-Palmerini. If you can get your hands on it it is a nice historical document. (See the introduction for download here: http://dingo.sbs.arizona.edu/~massimo/publications/)

I never interpreted Fodors defence of innateness in that discussion to be a joke, and I know of people that wrote Phd's on it, but perhaps they were all fooled, which is a real possibility. He is a jokey guy. But regardless of his humour, his starting assumption that learning concepts consists essentially of hypothesis testing on the world is what is wrong imo and as this is basically what underlies Cognitivism, so is Cognitivism.

The other reference is a book that might now seem a bit outdated but it is a nice book on (against) innateness written when connectionism was at its height: Jeff Elman and many authors, Rethinking Innateness.

Btw I think the definition of representation need not make any reference at all to the brain. I like the definition by Haugeland, which goes something like: A representation is an object that is used by a system as a stand-in for something else (presumably because the something else is not reliably present or accessible but needed nonetheless). Most representations are in fact external man-made artefacts, and it is my conviction that all representations are external man-made artefacts, and that we have wrongly come to mistake our own brain activity patterns for the kinds of things only complete living human beings (with brains in bodies in action), situated social settings, can produce.

Simon thanks btw for pointing us out to the Malafouris paper, I like it!

I didn't have time to read fully this manysided thread, but perhaps I may add my argument whý there are no "inner" representations (from https://www.researchgate.net/publication/226206085_The_theory_of_the_organism-environment_system_II._Significance_of_nervous_activity_in_the_organism-environment_system?ev=prf_pub): 

According to the theory of the organism-environment system, the basic principle of nervous functioning is not that of information processing, but creation of such constellations of neurons which - joined to the other parts of the body and environment - may achieve behavioral results which are useful for the metabolism of neurons and through this for the whole organism. The neurons are in many ways the most sensitive cells in the body and their large-scale destruction leads necessarily to the restriction of the action possibilities of the whole organism.

From this point of view it is clear that neurons do not create maps of the environment, inner models or representations which would somehow correspond to homuncular perceptions. Such reproduction of the properties of the environment in the nervous system is simply not important from the point of view of appropriate behavior, and must be assumed only if the starting point of the theory of nervous functioning is based on the absolute separation of the organism and the environment.

The necessary condition for forming systems leading to useful results is not from the systemic point of view that nervous organization should reproduce the organization of the environment as some sort of representation or model. The only essential is that a system may be formed in which elements belonging both to the body and to the environment are fitted together. The structure of the body, of course, "reflects" the structure of the environment in the sense that by inspection of the bodily structure we may also conclude something about the possible structure of the environment. When looking at the body of an organism we may speculate on what kind of environment would be appropriate. The study of the organism is simultaneously the study of the environment.

Let’s make our point clear with one further example. The system for cutting wood consists of a saw and a tree. In order to have a well-functioning system the properties of the saw and those of the tree should not be the same, but rather different in the way which makes a result possible. In a system consisting of two sets of elements one set need not to reproduce the properties of the other in order to create, as a whole, a functioning system. On the contrary, to have a good system for cutting the saw must be hard and the tree soft, otherwise the system will not function properly. The structure of the saw reflects in some sense the structure of the wood, but only from the point of view of the result (cutting).

If we think that the nervous system must somehow reproduce or represent the organization of the environment this would mean that the representations in the nervous system would always lag behind the events in the environment. However, if we think that the essential feature in nervous functioning is its fit with the environment this means that the environment and the nervous system have the same time. Perception, for example, is simultaneous with the object of perception in the environment. This means also that a stimulus does not precede perception, but perception is a process in which the fit of neural elements with the environmental events defines the stimulus. Therefore, reaction time is not the time for the processing of the stimulus, but the time for organizing the result

Hi Timo,

I perused the article you referred to and very much agree with your approach except for the conclusion that no inner representation is needed for a dynamic organism-environment system.

It is precisely because the system has the freedom to move it needs some means of representing its position to everything else including its own body parts. But your approach is logical and consistent because of  the larger scale of viewing – you look at the organism as a system within the biological environment. From this viewing scale the organism cannot be separated from the environment. However, this does not go far enough for me. In my research I adopted even higher viewing platform, that of thermodynamics, which emphasises the embedded nature of the organism even more. On this view, organisms are nothing more than thermodynamic processes (in the tradition of Prigogine) not much different to, let us say, a cyclone or convection current in a coffee cup.

However, if you adopt a lower scale for your observations, e.g. the scale of individual, no system can move voluntarily without some form of representation. The problem with representations is that they mean different things to different people. Most researchers think of representations as some sort of object, for example, a picture/painting the notion of which you invoked in your article. Whenever the argument about representation is brought up, it typically involves representing a real chair outside the body with the exact replica of it in the brain. Because this is absurd, the presence of a homunculus is postulated which leads to regression. Others avoid this problem by suggesting that the brain stores and manipulates (linguistic, mathematical, or program-like) symbols. But there is another much simpler explanation, a modified bundle theory described originally by David Hume.

Let us assume that the body is a self-contained system capable of moving in relation to its own body parts and, at the same time, to features of the environment. I hope this statement is uncontroversial to you. Moving in general involves changes to one’s own form which changes spatial relations to external objects. The only things we need to represent in the case of movement are spatial relations in the environment and the body. We do it in terms of various individual sensations which later are integrated so we attribute spatial relations within various modalities to the objects itself. This is referred to as sensory transparency which means that we experience external objects as a whole and not in terms of individual sensations. The sensory integration does not mean that we cannot experience those sensations individually. We do just that whenever we focus on one of the sensory properties. To cut the long story short, my approach is based on sensory maps as the only way of representing spatial relations within various sensory modalities.

Sorry for the long post. Wes

Timo,

thankyou for your post and paper. I am sympathetic to your holistic and stauchly anti-representationalist argument, (Though as Wes notes, 'representation' is a very slippery concept. I note that your paper was published in '98. The spirit of the paper reminds me of much of the embodied cognition literature, both before and since, going all the way back to cybernetic period, and especially the enactive cognition of varela, thompson and rosch. And of course more recent work by Chemero, Lakoff and Gallese, et al. I was surprised by the lack of referrnces to any of this work. It would help me understand the specifics of your position if it was juxtaposed with (any of) these others.

respectfully

Simon

Wes,

I like your metaphor of the organism as a cyclone; it comes pretty close to my concept of the organism-environment system which is not strictly speaking a system consisting of the organism AND environment, but an undivided whole (perhaps "living system" would be a better concept). This means, for example, that "movement of the organism" is not movement of the organism in relation to the environment, but rather reorganization of the system that makes different kinds of results possible. When "I move" to New York the organism-environment system is not "moving" to New York, but reorganized in such a way that as a result I may eat a hotdog on Broadway. Furthermore, from this it follows that internal-external dualism is abolished, and already there there cannot be any "internal representations". If you are interested how I apply these concepts to sensing you could have a look at the third article in the organism-environment series: https://www.researchgate.net/publication/225557038_The_theory_of_the_organism-environment_system_III._Role_of_efferent_influences_on_receptors_in_the_formation_of_knowledge

The only way how I would use the concept of representation is as the result of action of the organism-environment system. This may sound strange, as this sort of representation would refer to something in the future: it would be a "representation" of something that had not yet been presented. However, as the result of action of the system is determined by its history, such a representation would also relate to the past. Thus, "representation" would be an intersection of the past and the future; its description would be the description of the momentary state of the organism-environment system.

And Simon, during these years I have had many discussions with Chemero, Lakoff and Gallese, and I would say one of the main differences of our approaches is that they still see organism and environment as two systems. Varela et al. come in this respect closer to my approach as they regard organism as closed autopoetic system only "coupled" with the environment. A more recent description of my approach with some historical remarks may be found here: https://www.researchgate.net/publication/232840405_The_Theory_of_the_Organism-Environment_System_as_a_Basis_of_Experimental_Work_in_Psychology

Timo

thankyou for the link to the newer paper. I am pleased to see the mention of von uexküll. His conception of the world of an organism being given to it as a result of its particular sensori-motor formation seems close to your position, and merleau ponty reiterates this idea. But your position seems more radical. Have you read the enactivism vs extended mind debate by di paolo? His argument is that you can't be both extended and enactive because autopoiesis implies a subject/object division, ie  a self/world division. OTOH, EM, as wheeler points out, is 'a footnote to functionalism' ie it is fundamentally representationalist. Perhaps you have tried to explain this but ... how do you negotiate non-representationalism and denial of self-world dual- I'm not saying they are mutually incompatible. A resolution of this would be an interesting 'middle way'.

Simon

Simon,

perhaps it would be interesting for you to read an old discussion concerning my approach with Chemero, Clark, and others in Psycoloquy (see http://www.cogsci.ecs.soton.ac.uk/cgi/psyc/ptopic?topic=Efference-knowledge&submit=View+Topic ). Yes, you are right that my approach is more radical than those you mention, because I start with questioning the basic axiom of psychology (and brain research), i.e. the separation of the organism and environment as two systems. If you are interested you may also have a look at an article in which I describe the development of my thinking along these matters. (see  https://www.researchgate.net/publication/248904387_Some_background_and_further_theoretical_consequences_of_the_organism-environment_approach_A_reply_to_the_commentary_by_Panksepp)

Hi Timo, and others

Re: This means, for example, that "movement of the organism" is not movement of the organism in relation to the environment, but rather reorganization of the system that makes different kinds of results possible.

I find such views fascinating. They suggest that we all are parts of something bigger, without the need to invoke such inferior notions as tribes, nations or religions. For me at least, to see myself as a part of the larger and eternal cosmos is a humbling and strangely comforting view.

Note though that such a view relies on a single-scale perspective in which all interacting elements are of the same scale (e.g. think of cyclones). What is interesting about single-scale conceptualisations is that one could consider them in terms of either particles or atoms, cells or human individuals - where human individuals are conceptualised as being made of cells which are made of atoms, which are made of particles, and so on.

I find it fascinating that all single-scale ideas automatically invoke multiples of scales, and vice versa. Larger scales are created from elements of smaller scale by ensuring that those elements stay together while moving. For example, the living organism is a temporary collection of cells which developed ways to stay together as a single form that allows the collection to move in order to protect the integrity of its own form so it can move to protect its own form, .................

According to this view, any sentient collective-being can be seen as a particular set of spatial relations capable of duplicating (to some extent) and contextualising spatial relations existing in the environment in terms of its own constitutive relations, then changing the spatial relations that exist between the body and the environment in the way that ensures the perpetuation of its own constitutive spatial relations.

In simple terms, the sentient beings move so they can perpetuate the structure which makes this movement possible – where the structure and the environment is nothing more than spatial relations and the movement is just a change of those relations. The ultimate purpose of the movement (hence cognition) is to protect the integrity of the collection so it can move.

For all this we need some form of representation of spatial relations at the scale of the collective individual. 

My regards, Wes

Wes and Timo

I have been thinking about your recent posts and have some thoughts -

Timo, Thankyou for the recent paper. I am entirely in agreement with your desire to destabilise subject/object an self/world duals. I also note th spirit of your inquiry is consistent with Gibson, at least in the relational senses as interpreted by Chemero.

'a percept… is a result of preceding organisation' This is very provocative. Very interesting to learn of the work of Anokhin, thankyou! Regrettably, very little of his work is translated into english.

"A percept or emotion is not a representation located in the brain, but a result of organisation of the whole organism-environment system" Hooray for that, I say! … but it would still lead one (ie Panksepp) to ask - where and what is it ? (the percept or emotion). This is where the enactive/dynamical view is such challenge - we are not used to thinking in process ontology.

I absolutely agree with your instinct to query deep axiomatic assumptions, the self/world being such an axiom. Likewise your argument to query the intepretations of experimental results -and I would add, the theoretical frameworks from which those research programs emerge.

in relation to your opposition to the idea of 'the separation of the organism and environment as two systems': systems theory is a theoretical apparatus - it can be imposed at any level of scale - to use Wes's terminology -  for explanatory purposes. And it is always (?) directed at a particular process. If we are talking about, say, digestive processes, the relevant 'system encompasses subsystems of organs, microbiota, etc. And these are divisible into subsystems by the existence of distinct and discrete interfaces. I think what you are arguing is consistent with Haugelend's idea of mind embodied and embedded. I assume you know this paper (?)

When it comes to considering self-world as one system, I want to know the terms of reference. In terms of enactive cognition, I am entirely on-board. Yet Maturana and Varela impose a self/world dichotomy as axiomatic to their theory! I suppose we can speak of a larger system of digestion involving the growing plants and animals, but it seems to me we would lose some explanatory power.

Wes, your talk of multiple scales speaks to me of fractality, of dynamical systems and of the fundamentals of emergence theory. Your argument about atoms and cells and human and societies seems to me to confront the problem of non-reducability and multiple causality. Per the old adage, psychology is not reducible to biology, nor biology to chemistry, nor chemistry to physics.

Simon

Dear Wes and Simon,

thank you for the kind words. As to the multiple scales, I would like to share an exert from one of my articles (https://www.academia.edu/5374431/Consciousness_and_the_Ultimate_Essence_of_Matter):

"There are also other problems which seem to be very hard for the physicists: "How is it possible that a fundamental particle may be in interaction with the macroscopic measuring system ?".

This is a similar mistake which we have in the traditional psychology when we say that a stimulus causes perception. An element may be defined only in relation to other elements; an elementary particle cannot have any relation to the macro system, but only to other elementary particles. Similarly, man may not be defined as a human being among wolfs, but as a special kind of wolf. (A single human being cannot be related to the system of wolfs, but to the single wolfs). A "particle" loses its physical meaning without other particles with which it stays in connection. This connection is created in the measuring device. The result of measurement is, however, not a particle, but some new macroscopical state of the measuring device resulting from the interaction of the particles. This macroscopic state never interacts with particles which constitute it!

Classical objects, like measuring devices and slots are real in the sense that they may be included as wholes in our own scale of living. Non-classical objects, like elementary particles, are not objects in this sense, but only descriptions which our measuring devices have made possible of those aspects of reality which we may never directly observe. Correct physical theory may be developed only in the same scale. This seems to be the basic problem of ontological considerations of quantum physics where non-classical and classical objects are described in the same sentence. This leads to forgetting of metaphorical nature of descriptions of non-classical objects.

Similar mistakes are very typical of neuroscience, too. When we record activity from a neuron the neuron is at one end of the measuring system and the other end consists of the neurons of the observer. When the both ends (actually the whole system) act together we have, for example, an action potential which we may plot on paper or show in the oscilloscope screen and the every-day metaphor which we may then use is "spike", something sharp of which we think that it is transmitting information from one neuron to another, because we may see increase in the spiking frequency when we change the stimulus. There are, however, no spikes in the neuron, and there is neither any information transmission. These are also concepts for our every-day life and exist only for us, not for the neurons. The concept of a neuron is a similar abstraction as the concept of an atom: it is used as an explanatory concept to make sense of the results of some experiments.

Therefore, we cannot use pure neuronal data for description of human behavior as little as we may explain with the concept of atom why a table is a table. Human behavior is not realized by neurons alone as little as a table is created only by atoms. If there were no human beings the table wouldn’t exist as a table, because nobody would define the specific configuration of atoms as a table."

In conclusion, it follows that it is nonsense to say that neurons create images or pictures as representations of the "outside" world. Such "pictures" are the outcome of the complete organism-environment system as a consituent of a given culture.

Dear Timo,

It is the first time I came across someone who seems to share similar views. Your articles are the testimony of this. If the word “consciousness” in the publication you referred to is replaced with the term “cognition”, I would have no problem signing it with my own name instead of yours. I think you, just like me, must have felt quite alone and frustrated on occasions because of the unorthodox nature of our views.

Re: “This is a similar mistake which we have in the traditional psychology when we say that a stimulus causes perception. An element may be defined only in relation to other elements; an elementary particle cannot have any relation to the macro system, but only to other elementary particles.”

I would like to add a few comments.

First, let us talk about the issue most people cannot or do not want to accept – humans can only sense/detect the effects that ‘something out there’ has on their sensory systems or sensor-based apparatus. For example, when I close my eyes I do not experience the chair I sit on, only the effects the chair has on my body. Even when looking at my own body in the chair, I only experience the effects light has on my cones and rods. When I finally meet you in person (a big if) and shake your hand, I will not be experiencing you but the effects your body has on my body. In fact, we never meet our own children or spouses, not to mention other people – we experience them only as the effects they have on our bodies.

Second, the similar effect takes place in the case of instruments. Ammeters and pressure detectors do not sense the current or pressure, only the effect ‘something out there’ has on the specific structure of the instrument. Change the structure of the instrument (or sensory endings in the case of sentient organisms) and you will end up with a different ‘reality’. What is more, we do not experience instruments directly, only as effects their interfaces have on our bodies. So the “instrumental reality” is even further away from the reality of our own human scale (what I call “human reality”).

Third, when integrated, all those effects become transparent so we experience the object and reality instead of the effects of the object/reality on our bodies and instruments. We give names to those integrations (sons or daughters, particles and chairs, current or pressure, etc) and talk about them as if they and our bodies existed in the space of one scale. And yet we all know that a single particle cannot deform our bodies the same way the chair can. The particles and bodies are of different scales. And vice versa, the chair or the body cannot meaningfully deform the particle. Scales matter both in physics, in real world, and in human conceptualisations.

My regards,

Hi Timo, and all others,

I do not want to monopolise this site, but I think that the issues raised by Timo are essential for the better understanding of the question – is mental representation innate?

Regarding: “Consciousness was created in a system of several individuals when they joined their actions to produce common results. This joining was possible through communication which later developed to language.”

I view the role of languages in a similar but more general way. On my view, languages involve ‘physical objects’ created by smaller scale individuals in order to create larger scale collections (social groups). This is no different to human individuals building the infrastructure of cities (roads, factories, shops, churches) to bind them (the individuals) together into larger scale collections.

Different groups can be created by varying the objects with which the groups define themselves (compare the Bible to Koran or Das Kapital). The best objects bind individuals into very large and enduring collections and the lousy objects, like my theories and books, are incapable of attracting any individual.

The ResearchGate site is another example of a successful infrastructure which binds unconnected individuals into various groups. These ways of creating collections are no different to the way cells create larger scale collections (e.g. human individuals) by producing various connective tissues inside the space between them.

In this context, the notion of ‘physical objects’ stands for the ability of the objects to deform the collections in some effective way. So the spoken word “chair” is as much a physical object as the actual chair because it is capable of physically deforming some parts of the collective being. One difference between words and actual objects is that the words are much easier to produce (e.g. sound, writing) and store (books or records) then the objects they stand for.

Regards, Wes

Oops! I am sorry. It looks I killed another promising discussion .

:(

Wes 

Don't worry, Wes! I think you took quite an interesting aspect into consideration, which is also related to the concept of representation, viz. language. I certainly agree with you that words and "physical objects" are basically similar, and their use doesn't presuppose any "inner" representations.

From my point of view, an actual chair is "created" in a similar way as the meaning of the word "chair" when these parts of the world join the anticipatory organization of the organism-environment system in a certain culture. They are both outcomes or results of this preceding organization which gives them their meaning. An actual chair exists as a chair only for the human beings similarly as an inscription on the paper becomes a "word" only as a part of a capable reader. The difference between these outcomes is that they afford different actions: an actual chair sitting, and a word imagination.

Hi Timo, and all others

After reading some of your articles I am still unsure whether you accept organisms as distinct entities. It is evident to me that you interpret individuals as parts of some larger scale environments the same way cyclones are parts of the larger-scale weather pattern.

That much is clear, but I am confused because sometimes you suggest that the organism (and meaning) is literally created by the larger scale environment, which suggest downward causation, and on other occasions you maintain that it is created by the interaction between the individual and its environment. The difference between the two options is subtle but significant, at least to me.

In other words, to argue about a possibility of representations I have to know whether you are a follower of emergentism, holism, systems theory, or entirely something else. I favour the latest.

Regards, Wes

Wes,

perhaps this exerpt makes my position clear:

"It should be stressed that the theory of the organism-environment system is not a theory about how environmental factors should be taken account in the explanation of the behavior of the organisms, or how different contextual factors contribute to mental phenomena. The organism-environment system is not a system consisting of the organism and the environment which could be treated as subsystems of the whole system, but the organism-environment system is rather a methodological principle. This methodological principle entails that -- instead of looking at simple linear causal relations (e.g. the events from the stimulus to the response) when explaining behavior or subjective experience -- the research should start from the determination of the results of behavior, and lead to the necessary constituents of the living system determining the achievement of these results (see Jarvilehto, 1998a). The key concept of the theory is the concept of result which does not mean a simple effect or consequence of behavior, but a possibility of a new act, a transition from one act to another (see below). Thus, in the frame of the organism-environment theory, the starting point of all analysis of behavior is a historical and a developmental one. The organism-environment theory is essentially a theory of development, and its methodology is based on the analysis of different forms of the organism-environment system in their formation during phylogeny and ontogeny.

There are several important theoretical developments which bear direct relevance to many parts of the present consideration under such labels as "functional systems theory" (Anohin 1974, Shvyrkov 1990), "ecological psychology" (Gibson, 1979), "dynamic system theory" (Ford and Lerner, 1992, Port and Gelder 1995, Thelen and Smith, 1994, Clark, 1997, Hurley 1998), "theory of complex systems" (Kauffman 1993), and "autopoiesis" (Maturana and Varela 1980, Varela et al. 1991). However, their perspective is somewhat different from that in the present formulation, and there are also differences among them; it is the task of the future to relate all these theoretical developments together. There are also several "-isms" which seem to deal at least partly with the same problems and in a related way, such as emergentism, holism, general semantics, or postmodernism. For the present purposes I will only incidentally cite these approaches at points which in my opinion come closest to the present formulation."

If I have used somewhere expression "interaction between the individual and its environment" it is a lapsus lingue. As to the meaning generation during reading, I would say that an inscription exists as a word or meaningful unit only as it is integrated into the organism-environment system. There is no need for feature analysis or lexical search. Since inscriptions contain no information, they are informative only in the context of the organism-environment system’s anticipatory organisation. This arises from the subject’s history of social relations in a cultural context that, as a result, determines the unfolding of the anticipatory process. Every "word" is the result of the reorganization of the whole organism-environment system.

(see also https://www.researchgate.net/publication/232748964_The_role_of_anticipation_in_reading%2A)

Simon says:

"A percept or emotion is ... a result of organisation of the whole organism-environment system" … but it would still lead one (ie Panksepp) to ask - where and what is it ? (the percept or emotion). "

"I am still unsure whether you accept organisms as distinct entities."

I think in order to answer these questions we first have to be clear on whether we want to answer this 'within' the 'traditional' scientific perspective, or whether you we allow ourseves to see 'science' as just one, fairly recent, tradition of human practices, with its own 'way of understanding' the world. Science, in that case, is analysed within a larger framework of the phenomenology of our most basic experience of the world, which Heidegger sought to get a grip on. It is only in science that "organisms" exist as things that figure in theories of biology. We have grown used to using this word in our everyday language. Also the idea that 'everything', even perception and emotion, must always be 'somewhere' or 'generated by something' (a machine?) where the 'somewhere' is explicitly taken to be a position in extended space builds on the scientific assumption of a *reality*. All of this belongs to the world of objects that Heidegger calls 'vorhanden' (present-at-hand). This reality is not in opposition to the world of inner, mental states, and we need not choose between either physicalism or mental solipsism: it is precisely when we assume a physical reality that is 'outside of us' that an 'inner' world of 'mentality' automatically comes into being (Descartes). But before that is the more primordial raw experience that Heidegger calls Dasein - being-in-the-world and the first way of Dasein gets to the world in a Zuhanden mode: that of absorbed coping, dealing with the world without thematizing it as a set of objects in extended space.

I think - but this would be my question - that this is the world that Timo and Wes are trying to get at is this world.

Or do they wish to remain *within* the confines of regular science and (perhaps using dynamical systems theory and so on) wants to pose a 'scientific' version of 'organism-environment systems'?

btw Simon John Haugeland researched Heidegger for a large part of his life and wrote a book on his ideas.

Jelle,

good points; I think my theory is close to that of Heidegger (which I don't know much, although Sein und Zeit has been on my bookshelf for years), although I start from a different angle, viz. from the basic axioms in psychology. I have described the progress in my thinking in the following way:

"Finally, I realized that we could try to develop psychology in which we start with an assumption which is not obvious (in analogy with the concept of time in relativity theory), viz. that of the unity of the organism and environment. When I started to ask the old questions on this new basis it seemed that I could find solution to many persisting dilemmas. For example, my old question, the mind-brain problem, was unveiled as a wrong dichotomy: there is no mind-brain controversy, because the mind is related to the organization of the organism-environment system as a whole, and the brain is only one part of this same system.

However, I also realized that the new starting point requires that we must develop a completely new psychology and, furthermore, that this task cannot be accomplished in isolation from other sciences. We should consider how such systems have evolved in biological development, and we should also examine how to approach the social character of the human behavior. This sounded really like a "Mission impossible". Fortunately, I then realized that I was not alone, but several philosophers, biologists, and psychologists had already developed on a similar basis ideas in different fields into the same direction: Spinoza, Feuerbach, Brentano, Dewey, v.Uexkuell, G.H.Mead, Koffka, Gibson, Bateson, and many others."

See: https://www.researchgate.net/publication/248904387_Some_background_and_further_theoretical_consequences_of_the_organism-environment_approach_A_reply_to_the_commentary_by_Panksepp

Actually, the organism-environment system is not a very good term, because it could be understood as a system consisting of objects (organism and environment). This is not what I mean. Maybe a better description would be some kind of field, the reorganization (or turbulence if I continue with the hurricane metaphor) of which creates objects.

Dear Timo

I think I am very much thinking in line with your work. I like Von Uexkull, Gibson, Bateson. I will read into your work more thoroughly thanks for the links. I had been reading 'about' Heidegger for some years, relucting to actually read the work itself, but recently I started to read Z&T. I have now finished Division I. I like your rephrasing in terms of a field from which objects emerge. 

Working within design I became more and more interested in how all of this then ties into the social background/context/situatedness, that is also part of Heidegger but not always emphasized. I like the theories on social situatedness in the tradition of ethnomethodology and I think psychology can learn a lot from that field too.

And finally I am getting more and more interested in the idea of skill and how people produce artefacts and how these artefacts then become part of the 'being-in-the-world' of these people: Heidegger talks a lot about Zuhandenheit of the hammer, but he does not really distinguish well between the project that is created with the hammer and the hammer (and who was the person that created the hammer in the first place). He has this term Entworfenheit/Entworf, which in Dutch would be Ontwerp, which is in our language the word we use for Design. This intrigues me.

(But this is a bit off topic I realize)

Dear Jelle, thank you for your comment; it seems I should really read finally "Sein und Zeit". 

As to the original problem posed by Simon, I could summarize my position as follows:

If we start with the "two systems theory", separating the organism and the environment as two interacting systems, it is completely logical to postulate in the organism a representation of certain environmental features that will make the adaptive action of the organism possible. In this case I would say that such a representation is innate, at least in some rudimentary form that makes the survival of a newborn possible.

However, if we start with the one system theory, which postulates no organism without necessary environmental constituents, then the concept of representation is not needed, because there are no separate environmental features that could be represented somewhere. From this point of view, the question whether "representation" is innate or not, has no sense.  

@ Timo

I finally digested the recommended article so now I am in position to point out to similarities and differences between our approaches. The most significant difference is my use of physics (more specifically thermodynamics) rather than biology as the fundamental framework. Thermodynamics is also the source of similarities because it favours a single-scale systemic view of all phenomena and the arbitrary nature of divisions between sub-systems and their ‘environment’.

On this view the individual cell, its collections (the individual), or their collections (societies) could all be seen at the same time as distinct entities of different scales and as arbitrary regions of the overall system (the universe). What makes such regions distinct is that their respective elements move together in space. The galaxies, stars, planets and organisms exist temporarily as “bounded” collections of elements.

Consider organisms as an example. Even though they all can change their form, they are not entirely loose or accidental. When interacted with, organisms retain their internal spatial relations which can only be altered within the limits defined by the anatomy of the organism. These internal structures not only affect the way in which the organism can move but also how it can be deformed by other collections.

The two aspects are inter-related. On this view, the movement of an organism means changing various spatial relations between its elements (groups of cells) in such a way that the spatial relations between the organism and its “environment” are altered in a desired way (e.g. ensures integrity of the collection). All this requires the ability to sense and represent spatial relations both within and without the collection.

This also points to the importance of scales associated with collections of various sizes. All collections can be deformed meaningfully and non-destructively only by other collections of similar scale. For example, a chair is of the same scale as the human body because it can deform it meaningfully. Single cell, on the other hand, cannot deform the body, hence it is of a smaller scale. Similarly, the moon is of larger scale than the body because it would shift it as a whole without deforming it (only if you ignore gravitation in direct contact), and so on. The thermodynamic framework allows the researcher to move between scales with ease. 

The similarities between the two frameworks is, in my opinion, not accidental. Biology and physics are closely related. What is your view?

Best regards, Wes