Sleepwalk to Enlightenment

Illo credit "Anatomist90", over at Wikipedia

Illo credit “Anatomist90″, over at Wikipedia

Judging by the number of links I’ve received, a lot of you are already familiar with this paper on consciousness and the claustrum. Or at least you’re familiar with the tsunami of popsci coverage it’s received.

For the rest of you, the tale goes something like this:  54-year old female epileptic, seizure-free for four years at the cost of her left hippocampus. Now that reprieve has expired; the seizures have returned, and a team of neurologists led by Mohamad Z. Koubeissi have sown electrodes throughout her head to get the lay of the land and figure out what to do next. One of those electrodes edges up against the claustrum, a filamentous tangle of neurons thought to play a role in coordinating crosstalk between different parts of the brain.

When Koubeissi et al juice that particular electrode with 14mA of current, consciousness stops.

At least, that’s the way a thousand newsfeeds put it. More precisely, the body stops moving. The voice, which has been repeating the word “house” as a kind of baseline metric of awareness, trails off after a few seconds. The fingers, which have been snapping rhythmically, grow motionless. The patient sits glassy-eyed, to all appearances unaware and insensate. Inside her skull, the frontal and parietal lobes fall into mindless synchrony; not the synchronized call-and response of the consciousness state, but a mirrored lockstep incompatible with the operation of the global workspace.

Kill the current and everything return to normal. The patient reanimates, with no recollection of what happened during the down time.

The press is calling it a breakthrough.  An off-switch for consciousness, never before discovered. The Daily Mail, CBS, a myriad others have weighed in on the findings (although most of them seem to have mainly siphoned the bullet points off the New Scientist article that got there first). “…only a matter of time when we can create computers and machines that also contain a form of consciousness,” opines the Washington Post. “Their accidental discovery could lead to a deeper understanding of … how conscious awareness arises,” Discovery.com chimes in.

They keep using that word. I don’t think it means what they think it means.

No, the caption doesn't say what those asterisks are. I'm guessing, statistical significance?

No, the caption doesn’t say what those asterisks are. I’m guessing, statistical significance?

If I wanted to be glib I’d point out that a rock to the head serves as a perfectly effective off-switch for consciousness, and I’m pretty sure we stumbled across that result long before the latest issue of Epilepsy & Behavior hit the stands. It would admittedly be a cheap shot; after all, the claustrum effect is somewhat subtler. The victim didn’t keel over like a puppet with severed strings; she remained upright, eyes open, “awake but not conscious”.  That’s kind of cool.  And the claustrum’s involvement is nicely consistent with the whole Global Workspace model, the idea that consciousness somehow emerges from the integration of different brain processes talking to one another. It’s a good paper. The stats are solid, even conservative (although it would have been nice if they’d told us what those asterisks were supposed to represent in Fig. 1).

But closer to understanding “how conscious awareness arises”? I don’t think so.

What we have here is another neural correlate. Those are useful things to have, but all they tell us is that consciousness doesn’t manifest unless the machinery is ticking a certain way. They don’t get us any closer at all to the Hard Problem, which is: why does that particular flavor of ticking machinery wake up? When all those subcortical structures— the brain stem, the thalamus and hypothalamus, the ACG— start talking to the frontal lobes just so, why does it feel like this? It’s just computation, after all. Circuits in meat. Why does it feel like anything?

I don’t know if we’ll ever figure that one out.

I have other reservations. Prior to flipping the switch, Koubeissi et al got their patient to start repeating the word “house”, and to snap her fingers. They did this, we are told, to ensure that it really was consciousness that was being interrupted— that those milliamps hadn’t just induced some kind of motor paralysis that stilled the body even though the mind was active. K et al‘s reasoning was that paralysis would kick in instantly when the current hit; the fact that the speech and the finger-snaps trailed off gradually is supposed to take the paralysis confound off the table.

Yet there’s nothing in the paper to explain why this “off switch” couldn’t also activate instantaneously (once again, I cite my rock to the head). It seems a significant omission in the rationale, especially given that this “switch” has never been documented before. Besides, if the results had hailed from a conscious-but-paralyzed individual, wouldn’t she have been able to report as much after the fact?

Speaking of confounds, here’s another one. It wasn’t just “conscious awareness” that went down for the count; it was cognition.  The patient showed no response to stimuli during the vacant intervals; Koubessi’s team may not have induced unconsciousness so much as catatonia. (Interestingly, they also reported a “slowing of spontaneous respiratory movements” during the tereatment. This would seem to suggest that autonomic— i.e., nonconscious— processes were also affected. Unless the procedure itself was so stressful that the patient was breathing hard to begin with.)

Koubeissi et al unleashed a shotgun blast, insufficiently precise for high-resolution insights. This is no criticism; they weren’t performing a controlled experiment, just a routine diagnostic procedure that happened to yield valuable and unexpected results. But by that same token we should be careful about the conclusions we draw. (The fact that the patient’s brain was atypical— having lost half its hippocampus to a previous operation— has been dutifully noted in most of the coverage I’ve seen.)

What I’d really like to see would be a stimulus which shut down consciousness but left the cognitive and reactive circuits intact: a scenario in which the patient continued to repeat “house” while the current flowed,  until— still unconscious— she processed and accommodated a new request to start saying “yoga” instead. I’d like to see her wake up when the current stopped, look around, and ask in a puzzled voice, “Why am I saying yoga? I thought I was saying house.” Now that would tell us something.

What, you don’t think that’s realistic? You think consciousness and volition go hand in hand, that the body can’t parse the house-to-yoga transition without some little guy behind the eyes to make sense of it all?

I’ve got one word for you: sleepwalkers.

It’s possible to sleepwalk your way though a repeated series of sexual encounters with complete strangers (note to philanderers: don’t try this at home). It’s possible to drive across town and stab your  mother-in-law to death, unconscious the whole time.  “Homicidal somnabulism” is enough of a thing to warrant its own Wikipedia page.

So forget epileptics with pieces cut out of their brains. You want to find an off-switch for consciousness? Reserve the departmental MRI for the graveyard shift and put out ads for sleeping automatons. Some of them, short of spare cash, might just see the flyer some 3 a.m. and call you up.

Even if they don’t know they’re doing it.

FinnCon-05

This entry was written by Peter Watts , posted on Tuesday July 15 2014at 02:07 pm , filed under neuro, sentience/cognition . Bookmark the permalink . Post a comment below or leave a trackback: Trackback URL.

21 Responses to “Sleepwalk to Enlightenment”

  1. sleep walking is fascinating! When is the new website going to come up?

  2. Is it not possible, that, in the case of sleepwalkers, that rather than being entirely unconscious, it is, in fact, some alternate personality, similar but distinct from the original, pulling the strings during the sleepwalking episodes, only to be itself destroyed upon awakening, and subsequently resurrected on repeat incidents?

  3. Definitely an interesting and important paper, even if I’d agree with many of the reservations voiced above by Peter above. I wonder, though, if the talk of an ‘off’ switch is assuming the unity of consciousness in a realistic way? Definitely, sleep-walkers are less consciouss than people fully awake; however, that does not make them zombies (in the philosophical sense). I speak from personal experience: I used to be prey to the occasional nocturnal wander myself, and I can rememberbeing dimly conscious of what I was doing. The principal difference was the shitty, blunt-force-trauma thinking I deployed to achieve whatever it is I wanted. (Open wardrobe, full bladder––you get the picture.)

    Wistful anecdotes aside, I suspect the situation may be a lot closer to Semir Zeki’s account of microconsciousness, or the idea that there are multiple forms of independent consciousness that admit of being present in greater or lesser degree. The woman in the Koubeissi et al study would seem to contradict this, and that’s what makes the paper so interesting. However, it would take a lot more than one subject––and one suffering from a pathology at that––to make such a claim. Even at that, note the admission: Occasionally, the induced impairment of consciousness was associated with scanty, perseverative, and incomprehensible verbal output consisting of one or two syllables, with a confused look on the face.” Not consciousness as you’d like it, certainly; but is this unconsciousness?

  4. Re: the figure. Is there an accompnying supporting online material pdf? Maybe it would have more details.

    You could try emailing the corresponding author. That doesn’t always work*, but who knows.

    * https://www.youtube.com/watch?v=N2zK3sAtr-4 for humor and for realz appendix in http://reproducibility.cs.arizona.edu/

  5. Ah but are the sleepwalkers running on auto or just dreaming with the sleep paralysis disabled? We forget most dreams after all, but that doesn’t mean we’re experiencing them without conscious awareness (Though volition seems to switch only for lucid dreams. In my case lucidity switches on colour vision in dreams too, for some reason).

    Dream memories seem to be dropped in the recycle bin unless one makes an effort to take them out. So the sleepwalkers might be the same.

    That said, I recall once interacting with my briefly sleep-walking brother and the feeling I was talking to “someone else” was quite eerie and distinct.

  6. Continuing the Princess Bride meme, “Anybody want a Wernicke’s?”

    I’m most disturbed by the ability to hold a person in stasis and suspend memory processing electronically, but maybe that’s just me.

    http://www.pinterest.com/pin/42643527692438124/

  7. The patient had a case of epilepsy so severe they cut out a giant chunk of her brain AND IT DIDNT STOP

    This is NOT a generalizable dataset

  8. > I’ve got one word for you: sleepwalkers.

    Well, first of all, how do you know that sleepwalkers are not conscious? It is true that they have no memory of sleepwalking after they wake up (because the hippocampus isn’t working normally), but it does not exclude the possibility that they do have some minimal form of consciousness – abnormal and very limited, but consciousness still.

    >What I’d really like to see would be a stimulus which shut down consciousness but left the cognitive and reactive circuits intact.

    You might be interested in this quite fresh paper: http://www.cell.com/trends/cognitive-sciences/fulltext/S1364-6613%2814%2900104-1
    It’s a publication in the frame of Information Integration Theory – it gives a pretty good idea of what is and what isn’t possible without consciousness, and serves to show that consciousness has a very specific cognitive function – that is integration of higher-order information – meaning, more complex tasks cannot be done without consciousness.

  9. It is quite an interesting result. We’ll have to see if it can be replicated in other studies.
    As Peter mentions, it will be also interesting if some variant can leave some cognitive ability intact while rendering a lack of apparent phenomenological experience.

  10. I’d be interested in your take on the practice of meditation. On the one hand it’s all about enhacing consciousness, training to be “aware” at all times in the moment. On the other, it teaches that the self is an illusion and erases the barriers between self and non-self.

    And of course there’s all the religious baggage… but there does seem to be some research showing the practice isn’t just mumbo jumbo, it does have real physiological effects.

  11. Nestor:
    I’d be interested in your take on the practice of meditation. On the one hand it’s all about enhacing consciousness, training to be “aware” at all times in the moment. On the other, it teaches that the self is an illusion and erases the barriers between self and non-self.

    And of course there’s all the religious baggage… but there does seem to be some research showing the practice isn’t just mumbo jumbo, it does have real physiological effects.

    Yeah. It’s as though there’s something missing from this, though we are getting closer to an answer. While there may be some awareness of a problem being solved, as many writers will state, they don’t always even know exactly what problem it is that they are attempting to “solve.” Think Pete said he scrapped a short story because it wound up supporting the opposite of what he intended it to. Have seen Gibson say he doesn’t always know what he wrote means exactly until some time after. There is clearly an unconscious aspect to the work that one is not always conscious of, and, taken further, it is sometimes best to forget the process while performing a task. Thinking about your arm while hitting a ball is often a recipe for a bad swing.

    Meditation–and prayer–operate similarly: forget the problem, let something bigger worry about it. Does it matter what that something is from a pragmatic standpoint? Whether you think a car operates by explosions turning a crankshaft or tiny gerbils or elves, pressing the peddle still makes it go when the system is working.

    As Tom Regan says in Miller’s Crossing, “I don’t know. Do you always know why you do things?…Must be nice.”

  12. +1 for idea of microconsciousness or however you spell it.
    Makes sense that consciousness as we know it would just be an accretion of lesser consciousnesses.

  13. I was wondering, when I read the article on the news, if they hadn’t found consciousness but merely a failure point for the ego. (To crib the phrase from Eclipse Phase, because consciousness is to big a word to be throwing around regularly.)

    Either way, it’s an interesting discovery; either there is a circuit that can be directly attributed to the ego or, more likely in my humble opinion, there’s a failure point that essentially breaks the ego’s ability to function.

    To me, it makes sense that the ego is an emergent function of our brains instead of the intended one, a chorus of voices, some ancient and some barely formed, that somehow come together under the delusion of oneness. Maybe because the back and forth weighing of options takes place too fast for us to be aware of it.

    On sleepwalking, there’s another related (possibly?) phenomenon I’ve encountered- in the Army, we called it either “running on autopilot” or “the drone zone”. When sleep deprived, people do weird things and won’t remember them at all- until someone calls attention to it. I watched a guy in the mountain phase of ranger school dig a fighting position until his hands were bleeding. He only vaguely remembered the event when we snapped him out of it. The interesting thing is he was completely fucking coherent the whole time- “I’m digging a fighting position,” “Because we need a fighting position” etc. I wonder if the function of the ego is to select the applicable roughly autonomous behaviors from the crowded library of muscle memory and apply that behavior to the correct stimuli. (Because, of course, muscle memory and unconscious behavior will be more accurate than behavior that has to be thought about every time. Guys will go to the range in the drone zone, hit forty out of forty targets, then go back when they’ve slept for a while and are fully engaged and scrape by with a twenty five.)

  14. I’ve been wandering around a web of citations and related articles due to some article on lucid dreaming and delusions, and my trail has led me to an article by Metzinger, who people are already familiar with based on discussions earlier in the blog (The Ego Tunnel was recommended here, iirc).

    This is from 2013, open access.
    http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3868016/
    The myth of cognitive agency: subpersonal thinking as a cyclically recurring loss of mental autonomy

    Tangent…

    this is where I started. wondering about this paywalled article. (got a pdf from someone, thanks).
    http://www.ncbi.nlm.nih.gov/pubmed/25028078?dopt=Abstract
    Dreams, reality and memory: confabulations in lucid dreamers implicate reality-monitoring dysfunction in dream consciousness.

    then these two

    http://www.ncbi.nlm.nih.gov/pubmed/24021850
    Testing the involvement of the prefrontal cortex in lucid dreaming: a tDCS study.

    http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3369221/
    Neural Correlates of Dream Lucidity Obtained from Contrasting Lucid versus Non-Lucid REM Sleep: A Combined EEG/fMRI Case Stud

  15. Seruko (Free Candy Inside Van):
    When is the new website going to come up?

    We’ve got three new pages pretty much ready to go (we’d have four if I didn’t have to keep answering fucking emails and blogments). Still some issues with the graphics scaling as window size changes. Currently plan is to retire all the rifters stuff pretty much unchanged into Archive mode, and layer new stuff over top.

    As to when, plan is to coincide with Echopraxia‘s release. Whether we hit that target remains unknown.

  16. ShipOfTheseus:
    Is it not possible, that, in the case of sleepwalkers, that rather than being entirely unconscious, it is, in fact, some alternate personality, similar but distinct from the original, pulling the strings during the sleepwalking episodes, only to be itself destroyed upon awakening, and subsequently resurrected on repeat incidents?

    This is actually the premise of a story I haven’t got around to writing for about eight years now. My sense is that sleepwalking is a kind of dream state; and dreams are, in effect, “conscious” experiences in which the input is generated internally rather than being acquired through the sensory organs. So, short answer, yes.

    Lodore
    Wistful anecdotes aside, I suspect the situation may be a lot closer to Semir Zeki’s account of microconsciousness, or the idea that there are multiple forms of independent consciousness that admit of being present in greater or lesser degree. … Occasionally, the induced impairment of consciousness was associated with scanty, perseverative, and incomprehensible verbal output consisting of one or two syllables, with a confused look on the face.”Not consciousness as you’d like it, certainly; but is this unconsciousness?

    Okay, this whole “microconsciousness” thing looks fascinating but I won’t have time to read the actual paper until the weekend so I’m reserving comment for the mo. Consider this a bookmark.

    As for the confused look and the mono/duosyllabic responses— unless those syllables actually comprised something intelligible (“Help….me…” would be telling), I’d be tempted to write them off as the random motor discharges of a machine whose kill switch has a nonzero latency. “Confused look” might mean something, but not having been there, who knows?

    It all comes back to my comment in the post about being careful what conclusions we draw. This is a caveat which, I believe, covers my ass at least.

    Nestor: Ah but are the sleepwalkers running on auto or just dreaming with the sleep paralysis disabled?

    Again, dreams: conscious, but reacting to internally-generated stimuli.

    Something worth considering here is that dream consciousness is qualitatively different from waking consciousness; the dreaming self doesn’t seem to have any kind of critical faculties, for example. Your mother can manifest as a tree, your girlfriend as a biker chick with one giant rootlike hair growing out of her head, and you accept it unthinkingly. It’s only when you wake up that you realize how batshit crazy the experience was.

    Obligatory Cat Reference: cats, with certain suppressor circuits surgically severed, will leap up, stalk, and pounce upon imaginary prey while sleeping (at least we assume that’s what they’re doing; it’s obvious hunting behavior, anyway), while intact-brained cats will merely lie there sleeping, eyes closed, whiskers and paws all atwitch.

  17. Alek: Well, first of all, how do you know that sleepwalkers are not conscious?

    Sheila: Re: the figure. Is there an accompnying supporting online material pdf? Maybe it would have more details.

    There was a methodological reference to a previous paper, which presumably detailed the same technique. I didn’t follow up on it, because I wanted to rewatch “Rise of the Planet of the Apes” before checking out the “Dawn” thereof.

    Nestor: Ah but are the sleepwalkers running on auto or just dreaming with the sleep paralysis disabled?

    See above. With the cats.

    Nestor: Ah but are the sleepwalkers running on auto or just dreaming with the sleep paralysis disabled?

    Ditto. And did I mention the cats?

    Alek: You might be interested in this quite fresh paper

    Again, this is awesome. Will read this weekend.

    Alek: It’s a publication in the frame of Information Integration Theory – it gives a pretty good idea of what is and what isn’t possible without consciousness, and serves to show that consciousness has a very specific cognitive function – that is integration of higher-order information – meaning, more complex tasks cannot be done without consciousness.

    I dunno. Driving a car is a pretty complex task, and we know it can be done while sleepwalking. In fact, most of the time it’s done unconsciously even when we’re conscious.

    Nestor: I’d be interested in your take on the practice of meditation.

    We know, at least, that Buddhist meditation increases the thickness of the prefrontal cortex and right anterior insula (structures associated with attention, interoception and sensory processing).

  18. Peter Watts,

    the dreaming self doesn’t seem to have any kind of critical faculties, for example. Your mother can manifest as a tree, your girlfriend as a biker chick with one giant rootlike hair growing out of her head, and you accept it unthinkingly. It’s only when you wake up that you realize how batshit crazy the experience was.

    Have you experienced lucid dreaming? The key to the experience seems to be training oneself to actually notice precisely these things and become aware inside the dream.

    I’ve found the state to be maddeningly elusive, I can go months without achieving it for more than a few seconds – I tend to wake up, or believe I’ve woken up. But when it happens my dreams go from being in black and white to sudden vibrant colour, so it does feel distinctly like some part of my brain was being switched on suddenly. Some people find lucid dreaming easy to do, they probably dream in colour all the time, the lucky bastards… :)

    Richard Linklater’s movie, “Waking life” deals a lot with Lucid dreaming.

  19. Peter Watts,

    I dunno. Driving a car is a pretty complex task, and we know it can be done while sleepwalking. In fact, most of the time it’s done unconsciously even when we’re conscious.

    Let me reply by quoting the paper I linked before – http://www.cell.com/trends/cognitive-sciences/fulltext/S1364-6613%2814%2900104-1:

    “We suggest that, whereas some integrative processes can occur without awareness, their scope is limited to smaller integration windows, to simpler associations, or to ones that were previously acquired consciously. This (…) suggests that consciousness holds an enabling role in establishing integrative mechanisms that can later operate unconsciously, and in allowing wider-range integration, over bigger semantic, spatiotemporal, and sensory integration windows.”

    More about Information Integration Theory:
    http://arxiv.org/ftp/arxiv/papers/1405/1405.7089.pdf

  20. Nestor: Have you experienced lucid dreaming?

    Once or twice. Back when I was a kid I dreamed that a group of us were trapped in a rogue elevator car that shot out of the top of its hotel and started plummeting to earth from 50 floors up. I conjured up a little control panel that allowed the car to sprout wings, and piloted us down a gentle glide path to the ocean, where we splashed down, submerged (elevator car was no a dual purpose submarine), and went on a nighttime cruise on a coral reef. That was great.

    More often, though, the realization that I’m dreaming just strands me. Generally when I’m having a flying dream. At some point I realize that flying is impossible, and having lost faith, I’m suddenly unable to do it— which leaves me stranded on the roof of a building or a ledge halfway up a cliff, or something. Or, if the realization occurs during flight, I suddenly find that I’m not flying at all; I’m tangled in the high-tension wires running Calgary’s streetcar system, or I’m suddenly climbing a rope in some school auditorium whose roof and girders have been painted in a crude facsimile of sky and clouds and I just never noticed the difference before.

    Sometimes when I’m in mortal danger, the recognition that I’m probably dreaming give me the courage to leap right up into the monster’s face and embrace certain death. At such moments the brains tends to change channels, and I’m suddenly in another scenario entirely.

    Stupid lucid dreaming.

    Alek: Let me reply by quoting the paper I linked before

    Right. Thank you. I did download those papers, I just haven’t got around to reading them yet. (I will soon, though; they’re central to a project I’m working on.)

  21. I don’t think your house/yoga example would show what you think it would show. It seems to me you’d get exactly the same results if something disturbed short-term memory processing during the stimulation (or while it was ended), perhaps such that your short-term memory was more or less emptied (as when rising from sleep). If you were lucky and the ‘house’ part had got out of the shortest of short-term memory storage (the 7+/-2 chunk part), you could easily get this result regardless of what happened to consciousness.

    The Hard Problem — why does one form of computation in my head enact me, while other neural feedback loops in the same brain do not — remains unanswered regardless.