Self Comes to Mind
Constructing the Conscious Brain
Antonio Damasio
(2010-11-09). Knopf Doubleday Publishing Group. Kindle Edition.

At high level this book ponders the evolution of the information mechanisms of homeostasis, especially organisms with many parts. These were present in single celled organisms and evolved to be a significant part of our brain, and perhaps the ultimate purpose of all of the brain. Consciousness emerges as a form of mission creep in support of the extended phenotype.

Damasio quotes Feynmann: “What I cannot build, I cannot understand.” There are other notions of ‘understand’ and I am pleased that Damasio is at least aware of this notion. It is mine. There remains much yet to understand; many books like this will be necessary.

Warning: later parts of the book are in terms of the detailed anatomy of the brain. The books presents a hypothesis of where in the brain consciousness and the self reside. If you are overwhelmed by the neuroanatomy, jump to L 3838: “The Anatomical Bottleneck Behind the Conscious Mind”. Alternatively you may find the appendix (L 4549) to be an adequate guide to brain anatomy.

Two terms gave me considerable difficulty as I read the book. I think that it is not a spoiler to report here in my own words what I decided Damasio means by them:

Self
I think that by “self” Damasio means the memories that the organism (human) forms during its lifetime that influence its behavior subsequent to their formation. More abstractly it is that faculty of the brain that allows behavior to change as a result of experience—how we learn. I was badly confused on this point for most of the book and the confusion shows below.
Disposition
The hardwired (genetic) responses, such as reflex, to external and internal signals. The responses can have internal or external effects. Early brains were defined mostly in terms of their dispositions.

I have read only the beginning of Edelman & Tononi’s “A Universe Of Consciousness” and I hope to update this page when I finish that book.

Notes as I read

L 428: Damasio, in contrast to Edelman & Tononi begins with the body and how its data links to the brain. Damasio suggests that feelings are strongly tied to the body for perhaps so simple a reason that it naturally evolved that way. Once the body was the brain’s first job. One can argue that it still is.

L 448: It is clear that the ‘self’ that Damasio considers is not a member of the category of conspecifics; it is more elemental than that. My sore toe is very little like someone else’s sore toe; that comes much later. A few animals advance to making such a connection.

L 545: The ‘core self’ is here and now. The ‘autobiographical self’ is situated in time.

L 467: I think that Damasio’s ‘feelings’ are the ‘values’ of Edelman & Tononi. They influence the path of consciousness. They serve as a figure of merit.

I think that the ‘conductor’ is “attention”.

L 498: Damasio introduces “biological value” as a sort of economics concept related to and leading to the well being of the individual.

I agree.

L 529: Damasio begins to use the word “value” in a way that does not feel like either of the above meanings. It is sort of “what the consciousness mind want’s” which evolution might well push towards alignment with the biological value above. Perhaps the new meaning is like Edelman & Tononi’s which is invented as an explanatory mechanism.

I agree that these notions of value are useful in explaining the brain.

L 575: Chapter 2 begins with a long map between the logic of survival strategies of the single cell organism and of us. It is philosophical in nature but recounts ideas that I first learned from the anthology “Evolutionary Epistemology”.

L 795: Damasio introduces value (biological), again, and, gasp, price!. I am delighted.

L 1008: I think that there is a notion that should be introduced at this point. How is the self concept adaptive? I claim that the concept of conspecific comes first, and then that “I am one of those things.”. This allows useful generalizations on one’s conspecifics to be applied to the particular one that is of most interest—your self! I claim that you can be quite bright with no self concept—indeed you can be aware of external, and internal things and pay attention to outside activities without a self concept. I do think that self came soon after the concept of conspecific; it is however logically orthogonal. I may know that other people have feet and that I have a foot. It is a jump to notice that my feet are in a funny sense like other peoples feet. I doubt that any current robots (2013) have made that connection yet, or had it programmed into them. This is a bailiwick of mirror neurons.

I am pleased that Damasio thinks that being smart is adaptive. Some biologists seem to doubt this. I think you can be fairly smart without being self-aware but it is a drag. I think you can be conscious without being self-aware but this may not be viable. Perhaps you cannot be self-aware without being conscious.

L 1055: Chapter 3; Damasio uses “map” in the conventional sense of representation (in contrast to Edelman & Tononi).

L 1232: Here we review clinical recorded faculty loss resulting from loss of various parts of the brain. It is better organized that I recall elsewhere.

L 1433: Damasio speaks of “synchronizing” without explaining. It is in context of a 40 Hz signal. The computer scientist would assume in other context that a “clocked synchronous circuit” is in play. (See “Synchronous sequential logic” here.)

I am not ready to grant that the body contributes to emotion except via deployment of hormones which affect body and brain.

L 1590-1634: describes a proposed set of emotional activities naming several parts of the lower brain. The names of the brain parts mean nothing to me and I cannot imagine why he thinks as he does. Perhaps he will provide evidence later.

L 1678: Damasio proposes that mirror neurons that enable primates to mimic actions of others, are an extension of the brain’s understanding of what the body can do.

L 2064: The notion is that we evaluate other’s emotions, or our own emotions under hypothetical circumstances, using the same machinery as it we were having those same emotions for real here and now. This sounds like just plain good engineering—abstracting out the common functions. Perhaps it was not easy for nature to find, however.

This seems to define self only relative to others. Perhaps this is logically necessary. Perhaps an entirely unsocial entity has no need for a self concept. Self becomes a parasitic concept.

L 2070: confusion.

I take this opportunity to note that this book mainly reports the exploration of the brain from the subjective end; we have first class subjective knowledge of the meanings of the terms except, of course for “PMC”. I think we have no alternative but to proceed from both ends. The objective end speaks of brain regions, the subjective end of feelings and thoughts. It is like building a tunnel from both ends at once. That there is a physical brain locality in common to some of these subjective experiences is no doubt of eventual importance. OK but I am just a bit unhappy with the wording. I would say that the self is what is witnessed rather than the thing that witnesses. Perhaps it is not worth distinguishing between these two however; they seem to be almost the same. See the JTAG connection.
I must note here that as I read Damasio's book and listen to some baroque music on the radio that a part of my head is comparing Beethoven’s Große Fuge with Bach’s works and trying to decide what I would tell Beethoven about his place in musical history. I was not unaware of what I was reading; I recalled some of the details as I returned to Damasio after writing the above paragraph. Head-bones; knee-bones Proposed plumbing, or more accurately terminology: ‘Feeling’ is qualia depicting the emotions unless you exclude interoception from qualia. In either case qualia go to short term memory as they are perceived by consciousness.
There are a few distinct notions of ‘self’ I think. There are:

L 2262: I have trouble with Damasio’s phrase ‘dispositional space’. Google suggests that this is Damasio’s own phrase referring somehow to a higher level encoding of the sensory data. Some such encoding seems absolutely indicated. (See “color” here.)

This confirms that dispositional space refers to some sort of lossy semantic compression accompanied by an ability to decompress, but noisily. This is sort of obvious, yet an important hypothesis. It must be questioned, however, whether it is easy or necessary to translate backwards towards the original stimuli. I think it is not necessary and that subjective experience can be explained without such back-translation.

I find Damasio’s terminology obscure here. When I imagine how I would describe what I think he is saying, in engineering terms, it is still difficult but here goes: By semantic compression I mean that meaning in some sense is found in the raw data such as “That is a black cat.”. Here I refer to the meaning of that sentence rather than the words of the sentence. The meaning is captured in memory from those brain loci evolved to produce such meaning and fed back to roughly the same loci when the memory is recalled. By ‘meaning’ here I mean some set of neurons off and another set on.

L 2277: Damasio introduces the term “retroactivation” describing roughly recall of a memory.

L 2288: I need to understand Damasio’s “disposition”! Perhaps it is automatic (unconscious) reactions to stimuli, but then why not “reflex”? It seems to include ‘values’ which are, I suppose, like real numbers which the brain ascribes to many (most?, all?) frames. Another word for ‘value’ in other books is “valence”. A ‘goal’ of the brain is to bring about situations where some high value frame applies to the body. L 2146 seems to begin a principle discussion of ‘disposition’. L 2163 sounds like a definition that seems to me to be the same as a reflex except a reaction to some boolean combination of stimuli. In computer terms these dispositions amount to what a computer with ‘registers’ but no RAM could do. The dispositions do not rely on memory. In their ancient form the boolean expression was determined by the DNA. C. elegans would be an example. It is easy to imagine how such organisms would evolve so that a particular organism would change their actions according to previous experience of that same organism. That would be the first manifestation of memory. Such insights are likely to improve our understandings of the implementation of modern memory, but I expect that, unlike our eye, there are several separately evolved memory systems in our brain. Neurologists call the memory capacity of a neuron or synapse ‘long term potentiation’. It is illuminating to read the paragraphs of the book that include “disposition”.

L 2284: Damasio wants to segregate images and dispositions. I am OK with that but I think it might not be necessary. I think there may be a spectrum between the two. They are both denizens of memory.

L 2303: Damasio says that dispositions are not accessible but only deployable.

Good point. I would say that dispositions are merely that remembered potential actions downstream from our conscious attention—downstream too from some unconscious mechanisms.

L 2319: Regarding feedback signals in sensory brain areas: see this. Damasio uses other evidence (L 2381) from lesions to suggest that relatively up-stream neurons are employed to process visual images and these same feedback signals would be required for that too. I regard the text from 2375 thru 2400 to be good justification for figure 6.2.

L 2434: “Consciousness Observed”, I do not object to Damasio’s impressionistic description of consciousness that begins here. I may try to rephrase it to retain the logical content while avoiding the familiar and subjective words.

I take it from this quote that Damasio does not countenance a state of mind that is aware of it surroundings and reacts to them beneficially to itself, and yet is unaware of itself. I can imagine a conservation with such a being that would be unable to understand statements and questions with “you” in them, and yet be otherwise logical and friendly. I imagine that periods of time may pass in which the self notion plays no role in a conscious person. I suspect that Mozart would sometimes compose many bars of music while being hyper conscious and with no thought of himself.

Actually I think I misread Damasio. He goes onto exclude notions such as “entering the consciousness of a population” from his usage. The phrase “collective consciousness” may be useful but it is not what Damasio discusses in this book. Damisio’s consciousness is merely entirely private as is mine. (He veers off at L 2500.) There is at most one person associated with a consciousness, and probably at most one consciousness per person. (The last qualification is partly an artifact of some sort of mysterious unification tendency of the mind.) Language is a bridge, of sorts, between consciousnesses.

The ‘absence seizure’ described at L 2542 does not seem to me to be a lack of “self” but lack of ‘macro context’ which would include self. Not only the patient was missing from the patient’s world model, so was the doctor. I presume that the patient was not surprised to find himself in the doctor’s office and that some part of the brain retained that context. The patient also continued to act as if the doctor were not threatening. I agree that the episode is a powerful clue about the nature of consciousness but I am not sure why Damisio connects it with ‘self’.

I am curious how the patient would have responded if the doctor (Damasio) had said “That is my cup.”. The patient might have had a world model sufficient to respond. Perhaps saying “Please stand by the table.” would be simpler. I certainly understand why this did not occur to the doctor.

There is an important fact here but I am not satisfied with how it is expressed. I usually find myself digressing from the speaker’s train. This goes for either reading or listening to a lecture. Sometime two lines of thought proceed at once in my head. They are not entirely unaware of each other. Sometimes as I read I realize that another train of thought is transpiring in my head in competition to what I read, but not entirely displacing that. I test this hypothesis (of not displacing the printed text) by recalling at a shallow level what it said. When that happens I try to stop and decide which train to follow. Indeed such diversions lead to the bulk of this very page. The author goes off on a new tangent that I am unwilling to follow, yet. In conversations or lectures there is less latitude to veer off without losing the author’s thread to which you can return when reading. I do not believe this at least on philosophical grounds. There may be neuroanatomical reasons but I can conceive of conversing with an intelligence that is unaware of itself and indeed lacks any such concept. It carries on a coherent conversation in a context. Such an intelligence would be at a disadvantage but could be a contributor to society. It would indeed have specific blind spots. I think that you could introduce this intelligence to itself and convince it that it existed. Need Watson know about Watson? I imagine that Watson has an internal process a little bit like consciousness and it could learn about itself. I await evidence that Damasio needs this hypothesis.

L 2680: Damasio asks (for the first time?) of the purpose of consciousness. I think that we practically need to logically manage what we spend our time thinking about. We do this by ‘directing out attention’ we need a ‘thing’ to redirect. ???

L 2694: I am beginning to think that the Self that Damasio posits is the inducted ‘observer’ that observes the images recalled from memory and presented now by the eyes. If such images are real they must be useful and thus in some sense observed. It is not unreasonable to induce an observer.

This sounds like a fine description of the utility of the self. It co-evolved with building models of our immediate environment. The self was a thing to be included in the model in a special rôle. The self unaware intelligence whose possibility I posit would have no self preservation instincts and thus be unlikely to have evolved. It might, however, result as a mental aberration.

L 2751: Ah.. the entry of deliberation. That is a fine word for describing the process of raising hypotheticals (itself newly evolved) and considering which actions might lead to those.

Mathematicians may work for hours without even pencil and paper. They are sometimes distracted but more importantly they are sometimes not. Some claim that they cannot work in sensory deprivation but I think that this has not been tested. I cannot buy this. I have personally spent hours wondering about some math of physics problem quite consciously without involving ‘the material organism’ except as an occasional brief distraction.

Figure 8.1 [L 2818] leaves me cold. It divides ‘self’ into three categories which were presumably in evolutionary order:

Protoself
I can interpret his description as only the collection of signals from the body that are collectively presented to that portion of the brain that will evolve to consciousness. The purpose of those signals is presumably to inform the emerging executive of its domain. I cannot decode “neural description of relatively stable aspects of the organism”. Perhaps that means a model in the brain of salient organism state which varies on an hour time scale. More at L 2958.
The Core Self
Quote: “a pulse of core self is generated when the protoself is modified by an interaction between the organism and an object and when, as a result, the images of the objects are also modified”. The only meaning I can derive from this is that the ‘protoself’ is treated as a state and that the ‘core self’ consists of some sort of change in this state which is a result of sensory input. I presume that this state serves as a sort of model of the object. I do not understand how this relates to any normal notion of self. More at L 3122
Autobiographical Self
The fabric of memory of these changes. A person observes the world from a necessarily different perspective than another person. The ‘narratives’ are thus expected to differ. Is such a narrative the ‘self’?
xx It is clear that after evolution has given us a world model that include those nearby bodies that impact our well being, it is necessary to treat that body which is our own specially. That special body is one sort of ‘self’. I am still trying to grasp Damasio’s notion of ‘self’ and the above helps. This section (L 2802, “Building a Conscious Mind”) may be the principle discussion of self.

L 2863: Damasio begins a subjective introspection somewhat like Egan’s.

Evidently Damasio does not conceive of a being remembering a past except with a self. Evidently it cannot remember that it rained yesterday without a ‘self’. My hypothetical selfless minds are logically possible but perhaps biologically improbable.

L 2958: From this description of the protoself I can confirm that my selfless building manager has none and needs none. Some recent computer systems include a digital temperature sensor for the computer proper and this signal is led to logic that may turn off or slow down the computer to prevent damage. This might be incipient interoception. The brain’s early evolutionary function was to serve the organism. The building manager’s primary mission has nothing to do with itself.

L 3272: Regarding ‘autobiographical self’ it may be noteworthy that, at least according to Hollywood, amnesiacs forget their name, but not how to speak.

L 3475: I find this sections on “Sleep Research” and “The PMC’s involvement in the Default Network” opaque. A significant portion of the text is names of brain regions to which I have yet to assign function. What is said about these regions does not help me here.

L 3899: This section inspires me to define qualia. If you agree that ‘you’ are the functioning of some of your neurons, and perhaps a few stray chemicals, then you will agree that you become aware when someone turns on the lights. That is a causal chain from the lights to you. Qualia is where it gets to you.

I have a quick answer to the ‘should’ problem—valence. It informs the cognitive centers that this is a good (or bad) state and to plan accordingly hence forth. I am inclined to deny this as I worry about a mathematical problem. On the other hand I must admit that I am frequently distracted by visions of discovering an important theorem and perhaps motivated to proceed thereby. I am not sure if this qualifies under Damasio’s choir.

L 3933: Damasio seems to limit qualia to signals to the conscious mind of emotions. It conventionally includes sensory signals from outside the body.

I do not understand why Damasio wants to ascribe those admittedly vital ‘internal images’ to the self. I ascribe them to the organism, via its ownership of the brain. What has the self to do with it? The next passage: It sounds here like Damasio thinks that models were evolutionarily prior to brain mechanisms to preserve the organism and that the self came along to remind those mechanisms who comes first. (I don’t think Damasio really thinks that.) Models were never useful except for their protection of the organism.

L 4094: This section, “Self and the Issue of Control”, is very good and I think that in some sense I knew this at least as a teenager. I admit that many scholars wanted to deny this collaboration between the conscious and subconscious. Regarding culpability I always assumed that you put people in jail for what their brain decided to do, not merely because of what some sub brain decided.

L 4143: I am surprised that Damasio has not mentioned here the ‘confabulator’ module that seems able to find a logical reason for any action chosen by the committee.

Damasio makes clear here that memory of past images constitutes most or all of the self. In Damasio’s defense I note that memories distinguish the behavior of an organism from that of its conspecifics. (Some variation is genetic.) I see that ‘self’ can be thus applied. I have been wondering when Damasio would confront memory head on. It has been implicit in ‘self’ from the beginning. Damasion broaches here what I originally took “self” to mean—in short, self-awareness.

L 4400: I feel an itch to add to the consequences of memory: The organism with memory can adapt to circumstances too new for genetics to have captured—circumstances lasting longer than an interaction and shorter than the lifetime. I presume that long term memory is merely an evolutionary extension of shorter term memory.

I am 85% thru the book and Damasio’s title character ‘self’ still eludes me. Some of my notes here are written with my understanding of the word which clearly differs from Damasio’s. For me “self” refers to a concept in the brain of the organism upon which many other concepts, in the same brain, are parasitic. Damasio scarcely comes up to the level of concept in his analysis of the brain. I have not grasped another notion of self that a paramecium lacks. Perhaps the first informal attempt to introduce the word is here: A paramecium is ready for action. To seek any further reification of some mental process here seems to me to be only an illusion. But in this business theory must account for even illusions, but I don’t think that that is what Damasio means. Another passage: Well that sounds like my notion. There are a bunch of similar organisms out there with which the organism must cooperate or compete. There comes a time in the evolution of the organism that it must develop the category of ‘them’ and that it is indeed of that category. At that time it must be able to designate (with a mentalese noun) which one of them it is. The ‘self’ is then, finally, a necessary concept. The paramecium lacks this concept and all concepts, and all models. One might argue that since the paramecium sometimes behaves as if it had a model that we should metaphorically speak as if it did. Metaphors are good but the difference here is that the brain of the human has localized patterns that relate to ‘self’ where the paramecium has none.

I fear that I have not yet plumed the depth of Damasio’s meaning. I am beginning to think that Damasio has collected at least two things under the term ‘self’ which are better left separate.

I am disappointed that Damasio did not delve into how the brain processes the hypothetical which must have co-evolved with Damasio’s ‘self’ for it is how the brain evaluates possible choices in light of past experiences, and is also how the brain decides how some situation came about. Even the child distinguishes between the real and the hypothetical and it is a mystery how the brain accomplishes this.

(4549) 4631