Book Review: Consciousness Explained, (Danniel Dennett) (IX): Autostimulation

This part of Dennett's book attempts to explain the origin of "thinking". The hypothesis proposed is very interesting, very simple and convincing.


Consider a time when language just started to develop: hominids would have been grunting various sounds and start to attach meaning to them. Modern apes do that all the time. For a striking and well documented example, see vervet monkies's alarm calls for snakes, eagles and leopards.
We can speculate that at time, our early hominid would ask a question to his comrades by uttering something and could expect an answer (in a social group, "I scratch your back, you scratch my back" behaviours may be beneficial to the individual if  the group members have the necessary memory to keep track of other members behaviours). Our hominid ancestor could have developed the habit to ask questions and to answer other's questions (e.g. "are there crocodiles in this river?"). One day, one of our hominid ancestor might have asked a question while nobody was there to respond, but since he himself heard his own question, it triggers in himself the urge to reply to it! And to its delight, he found out that he just answered his own question! Asking oneself question could arise a side effect of asking other questions. Speaking to oneself could have an evolutionary advantage if the connection between the different parts of the brain was not optimal. E.g. a first part of the brain might need a piece of information present in another part of the brain, not efficiently connected (yet) to said first part. By broadcasting its request in the environment and by relying on a pair of ears to pick it up, said first part could establish a virtual link between both badly connected parts of the brain.

Talking out loud is one possibility, drawing pictures to oneself is another. One can also see how making such "communication with oneself" more private would be advantageous.


This scenario strikes me as a good candidate for how "thinking" evolved!

Consciousness Explained (D. Dennett): Part VIII; The Baldwin effect

We are the result of natural selection. Natural selction works by 1) allowing mistakes in the genome of an animal's offspring and 2) selecting the offsprings whose genome makes them better at survival/reproducing. If you are not familiar with this concept, go to buy "the selfish gene" from R. Dawkins (a live changing book).

The Baldwin effect is a mechanism that speeds up natural selection in a very peculiar way. It require you to have a plastic nervous system, in short: an ability to learn. If there is a "good trick" out there for an animal to learn, e.g. a behaviour that would greatly enhance an animal success at reproducing, any animal having this good trick hard-wired in its genome would be at an advantage. However, since having this "good trick" within it would be an all or nothing story, there maybe lacking a sufficiently progressive slope for natural selection to opearte swiftly. Hence, the time for this trick to be hard-wired in every animal within its species might be very long indeed. See left side of the figure below:

However, species not having this good trick hard-wired in their genome but having a plastic brain wired in such a way that they are capable to learn that trick during their livetime, are able to evolve the hard-wirering of this trick much faster. Here is how it works: Animal A has a brain slightly more able to learn the good trick than animal B. Animal A has therefore an advantage over animal B and will therefore probably leave more offsprings. Amongst said offsprings, animal A', due to a fortunate mutation, is still beter at learning the trick than its parent was, A' acquires it faster in his livetime than his siblings and therefore outcompete them and makes more offsprings. Amongst the kids of animal A', animal A'' is gifted with a further fortunate mutation that makes him particularly able to learn the good trick...and so on and so forth until the good trick (or at least the ability to learn the trick damned-easy) is practically hard-wired in the genome. What the Baldwin effect provides is a gentle slope for natural selection to operate on (see right side of the figure above). Any small improvement in the learning ability of the individual gives him a higher chance to hit on the good trick(s).

This makes clear the big advantage that a plastic brain provides to a species.

Book Review: Consciousness explained.Part VII. The evolution of consciousness

Daniel Dennett proposes the following evolutionary scenario for consciousness:

1) In the environment on earth before the appearence of life, some molecules acquired the ability to replicate (Richard Dawkins would call them replicators). This creates a point of view from which the world's events can be roughly partitioned between the favorable to this replication, unfavorable to it or the neutral".

2) "As soon as something gets  into the business of self-preservation, boundaries become important, for if you are setting out to preseerve yourself, you don't want to squander effort trying to preserve the whole world: you draw the line."

3)Next in line comes a certain ability to react to direct stimuli upon contact. E.g. engulfing food you touch or coiling away from harmful things you touch. This is the most rudimentary type of nervous response. It is a very crude way to predict the future. If I don't recoil, I will get eaten/destroyed/damaged.

4)Next, comes short range anticipations like the ducking reflex that permits to avoid being hurt be a projectile. Such reflexes are hard wired, including in humans (newborn infants have it). An interesting fact Dennett points out is the ability of many animals (from fish to humans) to be particularly sensible to the presence of a vertical axis of symmetry in its visual field. This is presumably caused by the fact that in our evolutionary past the most likely item in our environment having such an axis would have been a predator, a prey or a mate facing you. He argues that the fact that such alarm mechanisms are so crude, has the advantage of being fast and economical at the cost of many false alarms (the vision of a quasi symetrical tree for instance). For such a trick to be selected by evolution, it only has to give to its owner a slightly higher than otherwise chance to survive or to mate.

5) Once such a signal is detected, a further evolutionary advantage is achieved by a discrimination process capable to determine (at least in a crude way) if the first crude signal was a predator, a prey or a mate. In some fish, the vertical symmetry signal triggers an "orienting response", i.e. a swift interuption of ongoing activity. Dennett says that we have many such subsystems running unconsciously and performing specialized tasks.
When an alarm signal is recieved, e.g. due to the perceiving of symetry, our system stops and all our senses are open to maximize the input of information. If the alarm is confirmed, we get a rush of adrenaline and we react by e.g. fleeing. He says monst animals have such a system and he sees such "orienting responses" as a precursor to what we call our state of conscious awareness.

6) The animal acquired the ability to trigger such an "orienting reponse" also as a consequence of internal stimulis (not only external inputs such as symetric objects).

7) "Regular vigilance" turned into "regular exploration" where the animal frequently acquire information for its own sake, just in case it turns out useful in the future. E.g. primates with their saccaded eye movements scan uninteruptly their environment.

That's already a nice start I think...

Consciousness Explained: the contribution of a reader

I received an interesting comment from a reader (Jason Weber) that shed some more light on this book:

"I definitely suggest reading Sweet Dreams: it is very well written."

-I will order it…and read it in two or three years probably


"I took a look at your blog and enjoyed reading Part V of your book review."


-Glad to hear that.


"I have a question and a few comments if you do not mind. First off, what page did you find the quote for Dennett's alternative explanation for delay in consciousness?"

- p.164, 3rd§, line 2 : « Orwellian alternative »

"I ask because this sounds more like an "Orwellian revision" than Dennett's model for consciousness.
As I understand it, he does not support either an Orwellian or Stalinesque explanation, "...and the question Orwellian or Stalinesque? (post experiential or pre-experiential) need have no answer" (Dennett, Time and the Observer, http://ase.tufts.edu/cogstud/papers/time&obs.htm )."

-You are right that Dennett does not support either Orwellian or Stallinesque alternatives. I think I was wrong when I said that it is “his alternative explanation”. Actually he simply remarks that there is an Orwellian alternative to the Stallinesque model. Which I suppose makes his point that both explanations are possible, which is pointing to his multiple draft model.


"Thank you for bringing your blog to my attention, because I had forgotten how vehemently Dennett disagreed with Libet on the interpretation of those experiments."


-Strangely enough, I myself already read twice this part of the book and also forgot about it. Then I heard again about Libet type experiments and got quite excited about them, without realizing that I read a critique of them in Dennett. This really justifies my present effort to document what I understand from this book.


"Those experiments are very interesting to me and I had taken them at face value for a while."


-They are very interesting to me too. I am sympathetic to the idea that actions are decided before to be consciously felt as being decided…although I feel my confidence shaking a bit while reading this book.


"Your blog helped me to focus on some mistakes I had made. I do think that the unconscious parts of the brain plays a role in determining consciousness…"


-me too. I agree with Dennett that consciousness is probably distributed in the brain and that each location that get conscious, was not necessarily so the instant before. No single area for consciousness experience (I would have said otherwise six months ago. I might still change my mind ).


"…but that Libet's results and interpretations need a closer examination. Dennett provides that, so let me try to clear up exactly what Dennett does not like about Libet's experiments....

Daniel Dennett's criticisms of Benjamin Libet's famous experiment on intentionality:

In principle, I do not believe that Dennett disagrees with the notion that unconscious parts of the brain cause consciousness."-you are right


"He disagrees that attaining an absolute timing of conscious events from verbal reports is possible because consciousness does not happen in a centralized location in the brain.
That is the disagreement. Dennett notes that the representations of visual events occur, "...in various different parts of the brain, starting at the retina and moving up through the visual system...brightness...is represented in some places and times...location in others...and motion in still others" (Consciousness Explained, p.165).
Consciousness comes from many parts of the brain. For instance, he does not seem to believe that a single neuron or a single system in the brain possesses the state of consciousness all the time."


-It seems to me that it is indeed his position but I have a hard time understanding why the lack of a central location for consciousness prevents absolute timing of conscious events. Even if a conscious event appears distributed in space or at a location only temporarily made conscious, timing seems still possible.

"It follows that unconscious parts of the brain cause consciousness or change to the state of consciousness."


-indeed


"Dennett disagree with Libet because of his interpretation that the unconscious parts send there messages to a central conscious part of the brain."


-ok, I would need to re-read this part to check that it is indeed what Libet thinks and not Dennett’s interpretation of what Libet thinks. I think that even by abandoning the Cartesian theater, it is still coherent to interpret the results of the experiment by saying that the conscious “feeling” that a decision is made is felt after the decision. E.g. the decision is made due to a certain number of neurons being turned on at various locations…and 350 ms later we feel it because these locations changed “state” or maybe (my hypothesis) because this event “get burned” thereby creating a memory (at these same locations, e.g. by strengthening connections).


"Dennett states that, "...cognition and control -- and hence consciousness -- is distributed around in the brain, no moment can count as the precise moment at which each conscious event happens" (Consciousness Explained, p. 169). If measuring the absolute timing of a conscious event is possible, then we must assume that consciousness happens in one place and at one time in the brain."


-Not one place, only one time. No? Even if spatially distributed, it could be simultaneous or within a time relatively short compared to the delay observed by Libet.


"Dennett's model differs, not because unconscious agents cause consciousness but, because consciousness is not thrown together in one spot in the brain (e.g. the pineal gland). For Dennett, different systems contain different parts of the stories that make up consciousness but there is no place where representations come together to form a central conscious story of events."


-That is well understood.


"The most telling quote from Dennett in the chapter of Consciousness Explained that discusses Libet's experiment is this,

"Couldn't consciousness be a matter not of arrival at a point but rather a matter of representation exceeding some threshold of activation over the whole cortex or large parts thereof? On this model, an element of content becomes conscious at some time t, not by entering some functionally defined and anatomically located system, but by changing state right where it is: by acquiring some property or by having the intensity of one of its properties boosted above some threshold"

(p. 166)."


-I like this idea. It makes sense to me. I think consciousness intensity and memorization intensity are linked. Being conscious could be what it “feels” to memorize.


"The state change Dennett mentions is key: it is not that he believes that every part of the brain that has the capacity for consciousness is always conscious. I take it that he means that those parts change state from unconscious to conscious but do not do so until they receive enough clout to contribute to the distributed conscious experience. As different parts of the brain become conscious, the story changes; I think that is what Dennett means by multiple drafts."


-I think you are right and I think I now understand better this chapter. Thanks! I will re-read it nevertheless to be sure my new understanding fits with what he writes.


"Chapter six is difficult to understand so I hope I am properly representing Dennett's theory above. Feel free to correct me or to ask questions. Thank you,"

Time and the Observer

ase.tufts.edu

Two models of consciousness are contrasted with regard to their treatment of subjective timing. The standard Cartesian Theater model postulates a place in the brain where "it all comes together": where the discriminations in all modalities are somehow put into registration and "presented" for sub

Share

Here is part of a message of reply from Jason:

"...This might help to illuminate why Dennett is suspect of absolute timing of subjective experience:

In the last paragraph, I say that the "....state change Dennett mentions is key..." That is not exactly true. In fact, he warns that this view could lead us back to the Cartesian Theater, "...if it is claimed that the real...timing of such mode shifts is definitive of subjective sequence" (Dennett, Consciousness Explained, p. 166). His key point in chapter six of the book is that the, "...temporal sequence in consciousness is...purely a matter of the content represented, not the timing of the representing (Dennett & Kinsbourne, Time and the Observer, http://ase.tufts.edu/cogstud/papers/time&obs.htm ). Let's imagine an analogy to explore this point: mom is coming to visit you. Let's say that she decides Friday that she is coming to visit you on Monday. She sends you an email detailing her decision. Unfortunately, you do not check your email all weekend. Then, when Monday comes, you are utterly surprised to see your mom's smiling face when you open your front door. You then check your email and discover the news or your mother's decision to visit that day. The timing was such that you did not find out the news that your mother was visiting on Monday until she showed up on your doorstep. However, the decision was made and the message was sent earlier (i.e. Friday). This is why Dennett & Kinsbourne indicate that, "...temporal... details do not tell us directly about the contents of consciousness" (Time and the Observer, http://ase.tufts.edu/cogstud/papers/time&obs.htm ). Just because a message is sent does not mean we instantly become aware of it subjectively.


On your comment: "-ok, I would need to re-read this part to check that it is indeed what Libet thinks and not Dennett’s interpretation of what Libet thinks."
--I think you are correct that it is Dennett's interpretation of what Libet thinks.

Time and the Observer

ase.tufts.edu

Share"

-Thank you very much for your comments, they were helpful. Thank you also for accepting me posting your message on this blog.

Christophe

Book Review: Consciousness Explained (D. Dennett): Part VI:The Precognitive Carousel

Dennett reports an experiment made by Grey Walter. It consisted in presenting a carousel projector for showing slides to patients having electrodes implanted in the motor cortex of their brain. The patients were given a controller with a button for them to push on when they wanted to switch from one slide to the next. Actually, the controller was a dummy controler and the slides where changed in function of the signal received by the electrodes. The results? The patients experienced that the slide projector was anticipating their decisions. The slides would change just before they had decided to change them. This again seems to point to consciousness arising after that the decision is made. Dennett however once again express his view that timing of conscious events is not possible. an opinion of him that I still need to study further before to accept it or to reject it.

I received an interesting message on Facebook from a reader (Jason Weber). I will probably post it as a note to chapter V.

Book Review: Consciousness Explained (D. Dennett): part V : Delay of consciousness

In the 80's, Libet reported an experiment where subjects (wearing scalp electrodes) were asked to flex one hand at the wrist while noting the position of a moving spot at the precise time they formed the intention to flex. Afterwards, said subjects reported where the spot was at the moment they decided to flex their hand. The electrical measurements performed by the electrodes permited to determine that the subjects flexed their wrist 350 to 400 ms before the time indicated by the subject as being his time of intention to flex.


This experiment seems to indicate that we are conscious of the decisions made by our brain AFTER they have been made.


It seems to me  that Dennett does not like this conclusion (although he does not express it explicitly) which seems to contradict the intuition that our conscious acts control our bodily motions.

He considers this kind of explanation as being stalinesque (i.e. presuming that the brain delays consciousness in order to be able to presents all the facts it wants to present in the order it wants them to be presented).


He points out that there is an alternative explanation: "the subjects were conscious of their intentions at an earlier moment, but this consciousness was wiped out of memory (or just revised) before they had a chance to recall it". He calls this kind of explanation "Orwellian" because history is re-writen after the facts.

Finally, he also points out that yet another explanation which is that "an element of content becomes conscious at some time t, not by entering some functionally defined and anatomically located system, but by changing state right where it is: by acquiring some property or by having the intensity of one of its properties boosted above some threshold."

Dennett seems to doubt very much the possibility to equal the time of reporting of the subject with the time of conscious experience.


Personally, I would call on Ockham's razor with such an issue: stick to the simplest hypothesis until proven otherwise. Hence, as far as I am concerned, this experiment seems to indicate that consciousness arises after the facts. Clinical cases like the case of Mrs. Dee in "sight unseen" are also pointing toward such an explanation.


I think Dennett doubly dislikes the results of this experiment. 1) as mentioned earlier, it places consciousness after the events and 2) it looks like we are watching a movie of our live in a cartesian theater (he despises the idea of the cartesian theater).


[this page has been amended on 12/03 after a reader pointed to me some errors of interpretation. Thanks to Jason]

Book Review: Consciousness Explained (Daniel C. Dennett): part IV (his model of consciousness)

The main purpose of Dennett's book is to 1) refute the carthesian theater model of consciousness (see image below), and 2) introduce his own new model instead. He calls his new model the multiple drafts model.


The Cartesian Theater:


A representation of the carthesian theater model of consciousness is depicted below:



In this picture, we can see the head of a man looking at an egg being fried in a frying pan. Inside this man's brain, a screen is present where the vision of the man is projected for a smaller man to look at. The smaller man (or homonculus) is a methaphor for a central part in the brain where "consciousness happen".
The carthesian theater model assumes that their is a central "organ" in the brain where what happens is what you are conscious of. Its most naive form is the idea that a spectator must be present in the brain in order for the senses inputs to be experienced. Descarte for instance considered the pineal gland to be such a center from where the input of the senses could be transduced from mechanical signals (Descarte was not aware of the electrical nature of these signals) to spiritual meanings for the soul's benefit.


The Multiple Draft Model:




In this model, there is no centre where everything has to converge in order to be experienced. On the contrary, different conscious events are generated at different places in the brain. Each element within an event is discriminated only once. For instance, if you see a cow, the brain discriminate the presence of a large object, then it notice it is an animal, then that it is a cow. The discrimination and its fixation in the memory is enough for the phenomenum to be conscious (if probed).
What will be conscious will depend of the time of the probing. If you probe too early, you will only experience the diffuse presence of a large object. If you probe very late, you will have forgotten everything.
There is no need for the discriminated element to be sent or linked to a central theater. The conscious experience will originate from the locus of the discrimination. Also, each element is constantly updated/modified (e.g. interpreted, refined, partly erased, ...) due to interaction with the rest of the brain (pre-existing memories, new inputs, ...).
The multiple draft model makes"writing it down" in memory criterial for consciousness.


This can be best understood by looking at the color phi phenomenom at the bottom of this post (I do not insert it here because it is a moving image that would disturb your reading). To most people, the color phi phenomenom is experienced as a red spot traveling from left to right and changing color midway to become green. However (and you can easily convince yourself thereof by hiding one of both spots), it is in fact nothing of the sort. It is simply two fixed spots, one red and one green, the first blinking out of phase with respect to the second.


When experiencing the color phi phenomenom, the brain does not need to place intermediate spots after having experienced the second spot because there is no part of the brain present to "watch" these intermediate spots, your brain (i.e. you) just "knows" that the spot moved.


When remembering past events, you don't feel them happening again, you just know they happended. For instance, if you try to remember with as much details as possible the pain you experienced the last time you knocked your toes against an obstacle, you will not experience real pain, you will just realize that you "know" what kind of pain it was (which intensity, which location...).


His model makes a lot of sense but what remains unclear to me is the following:


At any given time our brain is processing a lot of inputs: external inputs from our various senses and internal inputs. We clearly do not feel conscious of them all. Cerrtainly, we do not feel conscious of many of them "simultaneously". Dennett seems however to say that whenever an object of our phenomenology is discriminated by our brain, it is "conscious".


On another hand he alludes to the notion of probing. Probing that would determine what is reported as conscious. He seems to make a difference between what is conscious and what we report as conscious experience when asked (i.e. probed) to report.


What I don't get is what he exactly means by his notion of "probing". I get it only partialy. I get it when by probing he means triggering a report from the conscious subject: e.g. asking the guy what he sees/feel. However, we are not constantly being asked to report on our internal states and we feel nevertheless conscious in these apparently "non-probed" times.


Of course, I suspect that his notion of what the probing is is more subtle than that. I suppose that internal inputs and external inputs can themeselves serve as "probes".


Maybe this notion is better explained in the rest of the book.


Once very important phrase he wrote is "The multiple draft model makes"writing it down" in memory criterial for consciousness".


I fully agree with the fact that writing something down in memory is critical for consciousness!


I have my own partial theory on consciousness and I should maybe explain it here before to continue Dennett's book. In view of the bold sentence above, I suspect that his views and mine overlap largly.

Book Review: Consciousness Explained (Daniel C. Dennet) (Part III: his method)

This is a continuation of part II

In his book, Dennett decides to rely only on what can be observed and recorded by an external observer of the conscious being. He does not wish to rely on introspection at all. Why this? Because he thinks (rightly to my opinion) that we are much less knowledgeable about our inner world than we think.

When asked, people will tend to think that their visual field is uniformly detailed and focused from the centre out to the boundaries. The following experience demonstrates otherwise:

Take a deck of playing cards and remove one card in such a way that you do not know what is on the card. Now close your eyes, turn the card so that it faces in your direction and place the card at the left periphery of your visual field. Now, when you will open your eyes, keep your focus on a point situated straight in front of you. Now open your eyes and, without stopping to stare at said point, try to guess which card it is. The amazing thing is that you cannot even say if it is a red or a black card! However, you can very well see that there is something there and if you move your hand a bit, you are totally seeing said something moving too. Now start slowly moving the card toward the centre of your visual field while keeping you gaze fixed right in front of you. You will be amazed to notice that the card really need to be damned close to the centre of your visual field before that you can even tell its colour or if it is a queen or king!

Book review: Consciousness Explained (Daniel C. Dennett) (part II: A party game called psychoanalysis)

I continue here my review started earlier.

This book is well worthreading can be obtained here.

  

3. A party game called psychoanalysis

Here, Dennett introduces us to a funny party game and argues that perception mights have the structure of this game.

The game works as follow:

1. Explain to the group that one person (the victim) will have to leave the group while another person (the dreamer) will have to explain his last dream to the rest of the group. Upon his return the dreamer will have to ask questions to guess what was the dream and who was the dreamer
2. Ask the victim to leave the room,
3. To the others, explain that the dreamer will have to answer all questions of the victim only by yes or no and according to a pre-defined scheme; for instance, the dreamer should answer "yes" if the question finishes with a vowel and "no" if it finishes with a consonent (or any other such arbitrary rule). Introduce also a rule, overruling the first one, imposing that the dreamer may not contradict himself,
4. Ask the victim to come back in the group and let him start asking his questions,
5. Once the victim "found" the dream (e.g. the dream was about a jaleous elephant with PZ Myers on its back entering the bedroom of the dreamer while he was playing monopoly with god) and guessed who had the dream, explain him how the dream was generated and explain him that said dream was actually purely the product of his (the victim) own brain and of a random process. Hence, joke that it is actually the reflection of the victim's own subconscient.

Dennett proposes the following parrallel between this game and the brain:

A part of the brain asks questions and these questions are answered on the basis of the data collected (by the eye for instance). By going back and forth in this process, the perception is refined, objects are identified, recognized, categorized.

For an hallucination to happen, all we need is to have the question asking part of the mind performing normally and to have a random or disorder or arbitrary sequence of yes and no as answers.

In the game, the questions asked by the victim's mind was supposed to reflect his current expectations, obsessions, worries,... Hallucinations are usually related in their content to the current concerns of the victim.

Hallucinations are the normal result of prolonged sensory deprivation. Indeed, the mind keep asking questions to the senses but get no answers, no input above the usual threshold for a nerve signal to be considered a valid input. As a result, the mind starts to lower the treshold until he gets some inputs. However, since there is no input at all but only the background noise, the mind need to lower the treshold so much that he ends up receiving said background noise. Hence, the brain recieves random answers to his questions.

These hallucinations start weak and grow stronger. This could also explain the origin of dreams...

What Freud´s theory of dreams says is that “something” in our mind is composing a dream for the benefits of our ego but disguises the true meaning of it. This, according to Dennett (and to me) is pure bullshit.

Book review: Consciousness Explained (Daniel C. Dennett) (part I: The brain in the vat)

Daniel C. Dennett is a proponent of consilience. He is trained as a philosopher and is the director of the center for cognitive studies at Tufts University. He is an interesting fellow because he is very knowledgable in sciences like biology or computer sciences and he applies his scientific knowledge and methods to the unraveling of the deepest mysteries such as ...in this book...consciouness. Not less!

I am only half through the book but I think that it can only do me good to summarize what I thought interesting so far.




Prelude: How are hallucinations possible?
1.The brain in the vat and 2. pranksters in the brain

If you have seen "The matrix", you know what is meant by "brain in the vat". Some philosophers have argued that it is not possible to tell whether we are really out there interacting with the real world or if we are just brains in a bocal filled with nutritive fluid and fed with inputs simulating a world.

Dennet expresses the view that we could be brains in a vat only if we were not given exploratory power. Give us exploratory power, even very limited, and the number of possible worlds that the vat master must generate for us to still believe in the illusion of a real life becomes enormous. Literally, it is a combinatorial explosion. In a nutshell, to get the feeling of the real world, you NEED the real world if you have exploratory power.

à we are not brains in vats

àstrong hallucinations are impossible (they are similar also simulations of a world we really believe in).

The credibility of an hallucination is inversely proportional to the strength of an hallucination.

Nevertheless, convincing, multimodal hallucinations are frequently experienced.

Dennet goes on to explain that triggering the optic nerve anywhere between the eyeball and the brain could produce an hallucination and that people having an hallucination are often very passive in the face of the hallucination.

To give the person the illusion of being active, the "illusionist" (e.g. the vat master) must know in advance the exploratory intentions and decisions of the victim or induce them.

Here stop my review of the beginning of Dennett's book. The following interesting guy seems to think that strong hallucinations are commonplace: