Metaphors We Daydream By

Neurophilosophy reports on how apparent motion affects mind-wandering:

Remarkably, it was found that the direction of illusory motion in the moving displays modulated the direction of the participants’ mental time travel. Those participants who had viewed the display with apparent backward motion reported that the daydreams they had experienced during the task consisted mainly, or solely, of memories of the past, while those who viewed the display with apparent forward motion reported thoughts related to the future. The displays used in the study produced an illusory sense of motion, so real movements could possibly have a stronger effect.

Ecological Realism in the History of the Concept of Time

Heidegger’s 1925 lecture course published as the Prolegomena to the History of the Concept of Time is widely considered to be the prequel to Being and Time. Since I consider BT to be a realist work in philosophy, it should not be surprising to find quotations in HCT that support what I call ecological realism. Ecological realism is to be decisively distinguished from philosophical realism. Both realisms agree, doxographically, that the natural world (what Heidegger later calls the Earth) would still exist if humans were wiped off the planet (Of course, this is just common sense, but many commentators seem content to ascribe to Heidegger the nonsensical position of anti-realism in respect to the ontic dependence of the Earth on human disclosure).But ecological realism differs from philosophical realism insofar as the latter accepts the “mind independence” of the Earth but arrives at this position from the starting point of a mind essentially cut-off from the Earth by means of the sense-data “veil”. According to philosophical realism, the only things we know directly are the sense-receptors at the edge of our bodies i.e. what’s called the proximal stimulus. Because they claim we are only in an epistemic relationship with the proximal stimulus, we must therefor “deduce” or “infer” that the distal stimulus exists. And moreover, because of the possibility of hallucination and illusion, we can never be certain that the proximal stimulus veridically corresponds to the distal stimulus. We are thus unable to adequately rebuff the radical skeptic of knowledge.

Ecological realism is different because it assumes that we have a direct epistemic relationship with the Earth by means of our intentional directedness towards the Earth. For ecological realists, the distinction between proximal and distal stimuli is a nonstarter in terms of epistemology. Indeed, Heidegger says in HCT:

When I perceive the chair and say, “The chair has four legs,” the sense of this knowledge according to Rickert is the acknowledging of a value. [However], even with the best of intentions one cannot find anything like this in the structure of this perceptual assertion. For I am not directed toward representations and less still toward value but instead toward the chair which is in fact given. (33)

Indeed,

[Rickert] is prevented from seeing the primary cognitive character of representation because he presupposes a mythical concept of representing from the philosophy of natural science and so comes to the formulation that in representing the representations get represented. But in the case of a representation on the level of simple perception a representation is not represented; I simply see the chair. This is implied in the very sense of representing. When I look, I am not intent upon seeing a representation of something, but the chair. (35)

For ecological realism then, the sense-data hypothesis is mistaken insofar as it begins with the assumption that representations get represented “in the mind”. For Heidegger, we need not make this assumption and indeed, we shouldn’t make it if we are to make sense of the phenomena of perceiving. One might reply by saying that Heidegger is merely assuming what he wants to assume in order to counter the neo-Kantians but why is he justified in assuming that intentionality is directed toward the environment and not toward the proximal stimulus? For one, the Heideggerian position is more parsimonious on the evolutionary and developmental stage because it allows for the possibility of coping behavior without the need for positing an internal consciousness “synthesizing” the proximal data into a percept of the distal stimuli, a processing heavy and thus energy-consuming process. If we look at the earliest progenitors of perceptual systems in unicellular bodies, we can see that the chemical receptors are directly connected with locomotion. The bacterium’s detection of instrumentally relevant chemicals sets off a causal cascade that eventually results in the spinning of the flagellum. In these bodies, there is no possibility of radical skepticism. The epistemic situation is akin to a gear that connects a car’s engine to the tires. Any possibility of perceptual mistake is thus physiological in character rather than epistemological. If we scale these systems up to humans, the same principle applies. We need not assume that the possibility of perceptual breakdown implies the possibility of radical skepticism. Understood from the ecological point of view, perception evolved so as to put us in intimate contact with the Earth. Ecological realism is also committed to the possibility of molar stimuli.

Moreover, in regards to the problem of other minds, philosophical realism assumes

that a subject is encapsulated within itself and now has the task of emphasizing with another subject. This way of formulating the question is absurd, since there never is such a subject in the sense it is assumed here. If the constitution of what is Dasein is instead regarded without presuppositions as in-being and being-with in the presuppositionless immediacy of everydayness, it then becomes clear that the problem of empathy is just as absurd as the question of the reality of the external world. (243)

By exposing the way in which philosophical realism depends on certain unnecessary assumptions about the nature of intentional perceptual systems, we can pave the way for ecological realism as both a metaphysical and epistemological doctrine. Metaphysical, because ecological realism accords with our common sense intuition that the Earth existed before humans and will continue to exist after humans are gone. Epistemological, because it offers a theory of knowledge in terms of opportunities of meaningful behavior.

‘[O]riginally and to begin with,’ one does not really hear noises and sonorous complexes but the creaking wagon, the ‘electric’ streetcar, the motorcycle, the column on the march, the north wind. To ‘hear’ something like a ‘pure noise’ already requires a very artificial and complicated attitude. (266)

This is of course an explication of our being-in-a-world with world understood in terms of significance and meaning. The world is directly meaningful because our affective care-structure compels us towards goals in a teleological fashion. This is of course much more parsimonious with evolutionary theory than any sense-data theory. There is much more to say on this issue, but I will leave the details for another post (and for my master’s thesis!). Also, In Jon Cogburn’s upcoming Fall graduate seminar, we will be exploring this issue of teleosemantics and animal cognition in great detail, so expect a flurry of posts related to these issues in the Fall. I can’t wait!

Light posting ahead

I’m travelling in Florida right now to visit family and friends, so posting might be a little light for the next week or so. I’m still getting a lot of reading done however, so if something stirs me, I might do some short posts highlighting interesting passages or making a few connections. I haven’t posted anything directly related to Heidegger in awhile, but I brought the History of the Concept of Time with me, so I might do a short post on realism if I am inclined.

Quote of the day: James on writing

There comes a time in all books when a man can’t tinker with them; he must write a new work altogether.

-William James

The Molar Stimulus: Implications for Visual Science

What is a perceptual stimulus? Many philosophers never ask themselves this question. They often take it for granted that vision science has settled the issue and nothing more needs to be said except that

1. Visual perception depends on light rays hitting the back of the retina, forming an “inverted image”.
2. The subsequent “retinal image” is imperfect and ambiguous in respect to “macro properties” since it is more or less “flat” , “upside down”, and registered by messy biological  tissue.
3. The brain has to “compensate” for the ambiguity by making “hypotheses” about the macroworld. The first stage of processing is the “2D primal sketch”. Next, the brain generates 3D “depth” from the more immediate primal sketch.  Thus, the brain is essentially a Helmholtz “inference machine” from the ground up.
4. Perceptual experience is what-its-like for the brain to make inferences. Visual experience is thus a simulation grounded by the immediate retinal registration.
5. Our intentionality is directed towards the simulated inferences rather than the world itself. Visual experience is thus illusionary and generated “internally” by the brain, hence the name internalism.

I take it that this crude picture is more or less an accurate representation of internalism. These presuppositions have become so entrenched, they hardly come up for review. I think it is high time to examine them, particularly in respect to the taken for granted assumptions about what constitutes a perceptual stimulus. The stimulus is often understood as being the immediate rays of light “bumping into” the retina. The retina is said to register the light strictly in terms of the wavelength and intensity i.e. the “primary qualities” of the physical sciences. Thus the retinal image corresponds to “micro” properties rather than molar properties. And because the registration of the retinal image is in terms of micro properties, molar properties like depth, surface, texture, etc. need to be “inferred” and are not directly perceived by the brain. It is said then that the micro properties are what make perception “ambiguous” and prone to error in respect to the macroworld.

But why should we assume that the brain is directed towards micro properties? It seems perfectly reasonable to suppose that the brain is behaviorally sensitive to molar properties specified in the ambient array of light “filling” normal environments. After all, molar properties are where all the action is. Only a programmed computer would be interested in micro properties. Biological bodies are far more interested in what’s going on at the molar level, where temporal events occur. Indeed, as J.J. Gibson points out, “Animals and men [directly] perceive motions, events, episodes, and whole sequences”.

And because all the episodes relevant to survival happen at the molar level (being attacked, hunting, socializing), it’s reasonable to suppose that the brain would have evolved so as to be directly sensitive to molar properties rather than micro properties. And if this is the case, then we need not assume that the brain is a Helmholtzian inference machine at the ground level of visual perception. Contemporary ecological optics has repeatedly demonstrated that molor properties such as depth, surface, and texture are directly specified by the invarient patterns of stimuli resulting from movement through the ambient optic array. The transformations across the retina specify molar properties in virtue of successive and adjacent order in the ambient optic array. Perception is thus capable of “holistic” sensitivity from the bottom up. Natural selection insured that brains developed the ability to be directly attuned to molar properties like motion, not micro properties like wavelengths and intensity. Motion indicates prey or predator. Wavelength indicates nothing unless you are a scientist.

We can thus overturn the five assumptions stated above. Intentionality is directed towards molar properties in the world rather than representations in the head. Because internalists are empirically mistaken about the ambiguity of perceptual stimuli available in the optic array, the absolute bottom-up inference making of classic computational approaches can be rejected as overcompensatory. Sure, the brain probably does make many inferences, particularly at higher-level stages in visual processing. But a complete Helmholtz machine? Unlikely. The availability of molar information in the optic array “ripe for sampling” obviates the need for a bottom-up inference from the micro to the molar. Our genetics ensure that nervous tissue is behaviorally sensitive to molar properties in the environment. This means that the “eternal now” of instantaneous registration is mythical and purely heuristic. Perceptual registration is temporally extended in virtue of its intentional directedness towards molar properties. We need to thus think about the brain in terms of a temporally extended coherence or “resonance” that is behaviorally sensitive to molar properties. This requires rethinking the constitution of perceptual stimuli and the evolutionary development of sensorimotor systems.

James on the Noumenal, a Reply to "Anti-Correlationists"

I have, indeed, said that “to be radical, an empiricism must [not] admit into its constructions any element that is not directly experienced.” But in my own radical empiricism this is only a methodological postulate, not a conclusion supposed to flow from the intrinsic absurdity of transempirical objects. I have never felt the slightest repsect for the idealistic arguments which Mr. Pitkin attacks and of which Ferrier made such striking use; and I am perfectly willing to admit any number of noumenal beings or events into philosophy if only their pragmatic value can be shown.

-William James, Essays in Radical Empiricism, p. 123

So called anti-correlationists would point to the last sentence and say “See! Reality is correlated with the pragmatic value of humans, therefore James was an idealist!” This response, of course, relies on a mistakenly shallow understanding of pragmatic value. When anti-correlationists think of pragmatism and “practical consequences”, they imagine using a fork to eat, or a lawnmower to mow lawns i.e. useful instruments. To say then that noumenal beings are only allowed in ontology when they have “pragmatic value” is taken to mean that entities only exist if they serve some useful purpose relative to human needs. But this is a mistake! In order to understand James, one must have an expanded notion of “pragmatic” that goes beyond mere instrumental usefulness. Imagine if the noumenal realm was so noumenal that Kant never even bothered to think about it, let alone write about it. Reflection on this indicates that the noumenal realm, whatever that turns out to be, must be cashed out in terms of its effects on us (what it causes us to do, even if that just means writing philosophically about it), otherwise we wouldn’t be able to talk about it or understand it. Notice how this is a purely methodological doctrine, similar to phenomenology. We need not directly experience these noumenal beings with our five senses, but they must factor into our cognitive economy somehow if they are to be discussed ontologically. Similar to the Kant example, if we postulate the existence of neutrinos that we cannot directly experience with our senses, the scientific discourse itself constitutes the “pragmatic value” of the neutrinos rather than any possible instrumental use we could find for the neutrinos. Thus, by understanding the broad scope of pragmatic value, which goes beyond mere instrumentality,  we can see how radical empiricism escapes from the “correlationist fallacy” often levied against James.

Benjamin Libet: A Jaynesian Interpretation

In 1980s, Benjamin Libet performed a psychophysical experiment on voluntary will that has sparked ongoing controversy regarding its interpretation and implications. Take the phenomenon of freely willing your hand to move. This is a phenomenon well-known to us all. Libet’s experiment was aimed at uncovering the extent to which unconscious processes precede the awareness of our decision to freely move. It had long been known that a “readiness potential” precedes voluntary action up to a second before the action takes place, a considerable amount of time neurally speaking. Libet’s reasoning for the experiment was as follows: if the conscious willing of the hand is causally efficacious it should precede or arise around the time when the readiness potential starts. Using a clever timing mechanism, Libet discovered that the readiness potential start 550 ms before the voluntary action and that the conscious willing is reported only 200 ms before the action. There is thus a “lag” or “delay” in consciousness behind the unconscious processing. Libet himself took this as evidence that conscious decisions for action are preceded by unconscious processing and that our traditional notion of the will as a commander is thus mistaken. All the will can do, says Libet, is “veto” an action. Libet concluded that volition is thus largely an illusion, with unconscious processing determining what we are going to do before we have a conscious awareness of making a decision.

But is that the only way to interpret this experiment? Surely not.  In fact, Julian Jaynes’ notion of a “struction” indicates that consciousness actually plays a powerful role in our mental economy, far beyond that of a “veto” function.  In the experiment, the conscious awareness of a decision was timed by a clock mechanism. The subjects were instructed to watch a spot of light revolve and report in clocktime where the spot was at the moment they felt they had decided to move spontaneously. Because the readiness potential always started 350 ms before the subjective decision to move (according to the clock), Libet concludes that the conscious will is not actually making the decision, but only capable of “vetoing” what the unconscious had already decided.

If look closely however, we can see a potential role for consciousness in terms of structions i.e. conscious instructions. Say you are the subject and you are tasked with making a spontaneous move and then reporting the time at which you had decided to move. One would likely narratize this goal in your head, perhaps rehearsing the instructions giving to you by the examiner. This would be the struction that guides your overall behavior in the experiment. Let’s call this conscious instruction to follow the experiment Struction A. Struction A would of course be operative in a temporally extended way throughout the experiment, otherwise you would not be able to obey the instructions of the experimenters. Furthermore, let’s assume that after making a conscious struction, the unconscious mind would need time to “obey” and “carry out” the struction.

This post-struction processing might account for the readiness potential and unconscious processing before the actual decision, but what about the 350 ms delay between the start of the readiness potential and the report of conscious will? Let’s assume that there would be two basic structions in the overall task. Struction A would be the metacognitive operator that guides the overall behavior so as to follow the instructions of the experimenter. Struction B would be the struction which actually gets reported as the conscious decision to move. Given that structions are brain processes, it is not surprising that brain activity would precede Struction B and that brain activity would follow Struction A. Moreover, given that structions are instructions from the narrative mind to the unconscious, it is not surprising that the actual implementation of Struction A takes processing time. But if you look at struction B without accounting for Struction A, it would certainly seem that conscious willing “lags” and never makes decisions itself (except to veto). But if we understand the way in which Struction A keeps the subject focused on the task at hand in a time-extended way, then it becomes clear that Struction A, in some sense, causes the entire half-second processing of the readiness potential, the conscious Struction B, and then the action itself. If you isolate Struction B from Struction A, it seems like consciousness cannot actually initiate decisions, but only “tags along” or gets “referred backwards” so as to seem as if consciousness is making decisions. But if we look at how the temporally extended decision to follow the experimenters directions keeps the subject on task so as to initiate Struction B, we can account for the “preliminary” unconscious processing of the readiness potential for Struction B without losing sight of overall place of conscious structions within the cognitive economy (guiding behavior at the abstract level of propositional attitudes through interiorized narratizing).

In summary, here’s what I think is going on. Struction A is made by the subject. This is a conscious willing, an “instruction” sent to the unconscious or subliminal mind for processing and execution. This instruction sends a command signal to the unconscious to prepare for the tasks of spontaneous decision making and the reporting of time. Unconscious processing takes place in the form of readiness potentials, which was catalyzed by the conscious command (which is experienced as interior narratizing). The unconscious processing initiated by Struction A sets up Struction B, the “spontaneous” willing, which is then reported as the actual spontaneous decision (according to the instructions of the overall experiment). If the subjects had been trained to extend their introspective awareness to include Struction A, then Libet would have concluded that consciousness actually does influence the action, albeit in a time-extended way through abstract executive control (acting and vetoing). But because Libet focuses the conscious report only around Struction B and not the total phenomenological experience, he failed to see how the unconscious readiness potentials which preceded Struction B were themselves preceded by Struction A, which was the conscious instruction to actually carryout the task of spontaneously making a decision.

The general lesson here is that conscious decisions are temporally extended because they operate over longer time periods that half second spontaneity. In fact, spontaneous finger-flicking is not a good model of how conscious volition works. For example, I can consciously instruct myself to raise my arm 10 minutes from now and ff course there will be unconscious processing occurring between the initial struction and when I raise my arm 10 minutes later. But it doesn’t follow that just because unconscious processing occurs before the arm raising that my conscious struction wasn’t the actual impetus for the action.

The Subliminal Mind

If the word ‘subliminal’ is offensive to any of you, as smelling too much of psychical research or other aberrations, call it by any other name you pleasure, to distinguish it from the level of full sunlit consciousness. Call this latter the A-region of personality, if you care to, and call the other the B-region. The B-region, then, is obviously the larger part of each of us for it is the abode of everything that is latent and the reservoir of everything that passes recorded or unobserved. It contains, for example, such things as all our momentary inactive memories, and it harbors the springs of all our obscurely motivated passions, impulses, likes, dislikes, and prejudices. Our intuitions, hypotheses, fancies, superstitions, persuasions, convictions, and in general all our non-rational operations come from it. It is the source of our dreams, and apparently they may return to it. In it arise whatever mystical experiences we may have, and our automatisms, sensory or motor; our life in hypnotic and ‘hypnoid’ conditions; our delusions, fixed ideas, and hysterical accidents, if we are hysteric subjects; our supranormal cognitions, if such there be, and if we are telepathic subjects. It is also the fountain head of much that ffeds our religion. In persons deep in the religious life, as we have now abundantly seen, –and this is my conclusion- the door to this region seems unusually wide open; at any rate, experiences making their entrance through that door have had emphatic influence in shaping religious history.

-William James, The Varieties of Religious Experience, p. 483-4

It is easy to see where James influenced Julian Jaynes. The bicameral mind is more or less James’ subliminal mind.

Thoughts on the Mind-Body Problem

If I asked an American layperson how he made sense of his subjective experience, he would probably tell me something about him “having” a soul, a mind, or a consciousness. If I asked him to point to his mind, he would laugh and tell me his subjectivity is not physical, but rather, “interior”, looking out onto the “external” world from “behind” his eyes. If I asked him whether he thought that his mind is the same thing as his brain-body system, he would tell me that the pure subjectivity of his phenomenal experience is an introcosm, nonreducible to the physical world. Accordingly, the layperson understands himself to “own” his body, controlling it like a rider on a horse. This is the famous Ghost in the Machine. The physical world is something our minds have to dominate and control, sending commands from the Master Control Room to the body like those cheesy sci-fi movies . How then to make sense of this mind-body relationship? The mind is mental and the body is natural. Can we naturalize the mind in the same way we naturalized the body? How can we make sense of the introcrosm or “interiority” that structures how we understand and feel ourselves to be? How am I inside my head?

This is of course the famous mind-body problem. It cuts a deep swathe in the history of philosophy. I can think of no other problem that has so much depth and goes by so many different names: reconciling the subject-object divide, naturalizing consciousness, intentionality, selfhood, subjectivity, solving the problem of knowledge, the problem of  the external world, the problem of qualia, etc. Theorists cannot understand what it  would mean to “explain” the subjective feel of “how it seems”, the redness of red, the coolness of water, the pleasure of sex. Philosophers are baffled when it comes to the mind-body problem; many have offered solutions but no consensus is ever reached, progress is never made.

But I want to claim that philosophers are stumped because they have a limited phenomenological understanding of subjectivity, not because the problem is too “Hard”. Failing to pay attention to the analogical structure of our self-interpretations, philosophers have mistakenly understood “mental interiority” literally, thinking that their “insideness” is real in the same way that an apple is real, but existing on another plane of existence, or at least property-ing in terms of “subjectivity” or “phenomenal feelings”. Most philosophers have failed to understand how exposure to inside-outside discourse in childhood analogically (i.e. metaphorically) constructs a “virtual” self-reflexive workspace that operationalizes whenever we jump offline and introspect on the perceptual gaze, understanding ourselves to be “feeling sensations” and subsequently narratizing alternatives to behavior through our working memory.

This phenomenon of navel-gazing is relatively new in evolutionary history. An animal on the savanna who took himself offline from his attentiveness to what’s going on around him (watching for predators, paying attention to social events, and looking for feeding opportunities) would likely not last long. Most nonhuman animals do not know what its like to “feel a sensation” because phenomenal sensations require the conceptual wherewithal to understand oneself as an “interiority having sensations”. Which isn’t to say that animals don’t have qualia. They just don’t have the qualia of knowing that you have qualia. There is a difference. And this difference makes a difference. Coming offline from the stream of perceptual sampling opens up new behavioral shortcuts through “chunking” world-information in terms of more abstract patterns. Narrative chunking is the great functionality of language. By “labeling” the micro in terms of the macro, we can overlook extraneous information, reduce algorithmic randomness, and streamline our behavioral affordance-space. Names, for instance, greatly reduce the computational load in social cognition. They allow for the possibility of personality ascription and more accurate simulations of behavior in accordance with our memories. Labeling works the same way. Seeing objects as objects allows us to point things out with great precision, transferring information and meaning with greater efficiency, increasing the efficacy of childhood training. Because forgetting information is so costly (storing it is the easy part), linguistically driven cognition opens up dramatically useful offline thinking through an affordance-space of streamlined behavioral alternatives. We can “narratize” different alternatives in behavior and use this narratizing as a new control schema for behavior. We can literally “tell ourselves what to do” (or submit to the authority of Others, as is usual).

Moreover, these “epistemic behaviors”, like all behaviors, have unique phenomenal profiles. But the phenomenal profile is structured by the narrative scaffold itself, and the mentalistic discourse of each historical epoch self-reflexively shapes the phenomenal feel of narratizing. Because our neolithic ancestors did not have inside-outside discourse and could not think of themselves as subjective minds “inside their heads” experiencing qualia, they never had the analogical foundation for constructing the offline linguistic workspace that “chunks” and therefore understands behavior in terms of propositional attitudes such as beliefs, desires, passions, etc. Only through childhood exposure to reason-based discourse can we utilize the implicit semantic structure of the logical space of reasons as a cognitive foundation upon which to construct an offline workspace that chunks the world in terms of mentalistic narratives. Through exposure to narratives, the mind can “bootstrap” itself into a dual-channel system, capable of nonconscious online behavior and offline workspace narratizing. The former allows for habits, the latter for thoughts.

We thus have a rough answer to the mind-body problem. We must first not take our own phenomenal experience literally. We need to understand that without our exposure to reason-based narratives in childhood, we would not experience ourselves in terms of mental states mappable in terms of psychological laws involving propositional attitudes. This is what the Ancient Greeks discovered. They understood better than anyone the possibilities of thinking opened up by language, but they too failed to take a properly meta-meta-cognitive stance on thinking because they (1) did not have the metaphorical models necessary to understand thought nonliterally and (2) they had no concept of the nonconscious mind. If we don’t take phenomenal feels literally, then solving the mind-body problem becomes a matter of explaining analogical construction mechanistically. I’m not sure what figurative understanding would look like computationally, but one thing seems clear: you need to have a body of some sort. If we could make an embodied computer learn analogically, we might make the Myth of Jones a reality.

Shifting Paradigms: An Interesting Conversation with Eric Thomson

Eric Thomson and I are currently having a stimulating conversation over at his blog neurochannels. It seems that we got ourselves into a debate about some of the very fundamental issues of the mind sciences, including the nature of perceptual stimuli, representationalism, behaviorism, computationalism, constitutive conditions of consciousness, dreaming, hallucinations, etc. I find this interesting because it represents a Kuhnian “crisis of foundations”. Eric does an excellent job of representing the classic computational paradigm currently dominant in neuroscientific circles. I attempt to represent the underdog 4EA theorists, who say that instead of the mind being a wet computer, the mind is embodied through our bodily  interaction with the  world, embedded in the social milieu, enactive rather than computational, extended into the artificially modified props that we use on a daily basis (clothes, computers, cooking, etc.), and radically affective in being grounded through the bodily aesthetics of emotional processing.

Eric is quick to point out that 4EA approaches seem unable to account for states like dreaming, hallucination, etc. i.e. states where there is experience, but it isn’t grounded in embodied motor movements. My reply would be that we need to explain higher-order phenomena like dreaming from the ground-up. This means that the subpersonal behaviorial loops emphasized in 4EA theories become the “substrate” or “foundation” for more complex experiential phenomena, such as dreaming or illusions. If it wasn’t for the experiential substrate of subpersonal processing, there could be no basis for thought itself! For “who” would the thoughts be relevant to? A ghost? Without a body, thoughts remain groundless.

Talking to Eric has made me realize that 4EA theorists need to do a better job emphasizing that we are not denying complex phenomena such as dreaming or propositional thought. On the contrary, we are trying to explain them as grounded by the sensorimotor substrate. 4EA skeptics seem to think that embodiment theory requires we explain dreaming and thought in terms of behaviorism, as if we somehow just forgot about the failures of behaviorism. While some 4EA theorists might be this reductionistic, I don’t think embodiment theory entails any such flat ontology. Instead, 4EA theorists should be willing to talk about, as Andy Clark does, dynamic on-the-fly online processing and “offline” processing. How the latter get cashed out remains a significant problem, but we shouldn’t be scared of using terms like “construction” or “scaffold”. As long as we realize that the low-level sensorimotor experiences themselves aren’t constructions “in the head”, we should have no philosophical difficulty explaining how it is only the the high-level states which become constructed “internally” by means of some sort of virtual workspace.I think George Lakoff and Mark Johnson’s work on inner-outer container schemas are really relevant on precisely this issue of internal construction. Metaphor theory provides the conceptual bridge to cross from body to mind, from online to offline.