Blindsight and Consciousness
Thomas Metzinger is a philosopher who pays lots of attention to cognitive science and psychology, and likes to think about consciousness. Most of the interesting ideas that follow come from his books The Ego Tunnel and Being No One. I hope to write a series of posts summarizing some of the evidence and arguments in Being No One, which focuses on consciousness.
Blindsight1
Blindsight patients have damage to their primary visual cortex (V1), leading to a scotoma, or area of blindsight in the visual field. Most but not all visual signals go through V1, so they can still influence the brain in very restricted channels. Blindsight patients don’t report seeing things in their scotoma, and don’t initiate plans based on it. If they’re thirsty and there’s a bottle of water in their scotoma, they don’t pick it up and drink it.
Human subjects and animal subjects are treated differently in psychological experiments regarding what they do and don’t know. Humans are generally asked to report on their own experience, while animal actions are observed. We get interesting results when we ask people to report on their experience, while also observing their actions.
If you ask a blindsight patient what they see in their scotoma, they respond to the point that they can’t see anything there. However, if you tell them to do things like “grab the thing in your scomata” they can grasp it. If you ask them to guess what’s in it, they can perform better than chance. Some blindsight patients can tell if something is moving in their scotoma, but they can’t tell you what it is. They often describe this awareness as a hunch.
Most people consider it fair to say that blindsight patients are not conscious of the things in their scotoma.
Attention and Conscious Experience
Patients with blindsight can act on visual information in their scotomas in some ways, but they can’t notice it.
Metzinger argues that humans don’t have a conscious experience of what we can’t pay attention to. Note: There’s a difference between can’t pay attention to, and not currently paying attention to.
Visual information in the scotoma isn’t accessible to the parts of my brain that plan, or the parts that cause me to say “I can see X”. My unconscious is able to refer to this information for things in forced choice situations, but the information isn’t available to me.
Constraints on Theories of Consciousness
Any theory which says that you need to be conscious in order to do things is probably wrong. Also, robots work. And machine learning exists. See also unconscious goals.
It’s possible for your brain to refer to something, but not have it be consciously available to you. It’s also possible to change what these things are.
The parts of your brain causing you to say that you notice something can be cut off from the parts that let you do things. This implies that some neural processes lead to you being conscious and others don’t, and that those processes can be interrupted without ruining everything.
Citations, Notes:
1″The Case of Blindsight” by Weiskrantz in the Blackwell Companion to Consciousness (you can get it here, though there are other place on the internet that talk about blindsight)
Heavily drawn from The Ego Tunnel and Being No One (both by Metzinger).
Thanks to John Salvatier for reviewing drafts of this post.
I once tried reading Being No One (after I saw the recommendation in Watts’ Blindsight), but I couldn’t get anywhere with it—it seemed way above my level. How are you copying with it?
His book The Ego Tunnel is waaaay better written/more accessible, and has helped a lot with going through Being No One.
I’m also reading the book in a weird order, (chapters 1, 4, 2, 3, 6, 7) and taking notes.
Way above your level in what sense?
I felt I lacked the biology for it, and I wasn’t sure I had the philosophy for it either. I simply couldn’t follow even the overviews.
It at least implies that some processes lead to you being conscious, while others don’t. The same brain region could be involved in both conscious and unconscious processes.
Even many of the same processes might sometimes lead to you being conscious and sometimes not. (the difference might involve the level of communication between different parts of the brain, which could be influenced by all sorts of things)
Fair point, corrected.
On a barely related note, I highly recommend Peter Watt’s hard sci-fi take on this topic http://www.rifters.com/real/Blindsight.htm—it was a lot of fun.
You seem to be implicitly equating “X is conscious of Y” and “X can talk about Y.” I’d recommend either making this equation explicit and justifying it, or clarifying your language to eliminate the implication.
It’s more subtle than that. It’s probably even more subtle than the following: If X can talk in general, and X is not currently under the influence of anything that prevents talking, then if X is conscious of Y, X can talk about Y. More flexibly still, the ability to talk about Y is the leading indicator of consciousness-of-Y.
Sure. It might be easier to approach the question this way: can you unpack your reasons for believing that if I talk about Y, it follows that I’m conscious of Y?
There isn’t much to say. Cases in which people talk about things they see, hear, think about, etc. are paradigmatic use cases for learning the phrase “conscious of”. It doesn’t follow (as a logical deduction) that you’re conscious of that of which you speak, but we’d need a really good reason against, to cast doubt on it.
There also have to be words relating to Y in a language X knows, and X has to know those words and what they mean. (I run into this issue sometimes with my synesthesia, which occasionally generates ‘impossible’ colors and shapes.)
Well, the question arises of whether a phrase like “that color my synesthesia occasionally generates, for which there isn’t really a word in English” qualifies as talking about the color. I would say it does.
It does, but the fact that there are cases that are borderline with regards to being able to talk about things implies that there might be cases that are on the other side of that border.
It may be more obvious with some of the ‘impossible shape’ cases—I can’t describe those at all, but I can say that a particular ‘shape’ is visible in both concept X and concept Y and that it relates to trait Z. (This is mostly interesting in practice because ‘seeing’ the ‘shape’ is often the first way I notice that a concept has a particular trait—for example, I recently noticed that ‘colorblind anti-racism’ is often a form of belief-in-belief because I was considering the concept and noticed that it shared a shape and color with other instances of belief-in-belief.)
So, I agree that some things are easier and harder to talk about, and that the existence of something X hard to talk about is evidence in favor of things X+epsilon hard to talk about. But I don’t quite get what real thing the metaphorical “border” corresponds to here.
Hm.
Perhaps I’m conflating ‘talk’ and ‘communicate’ in a non-useful way. A significant part of my point is that I can’t communicate these qualia—there’s no way to render green-and-purple or yellow-in-grey in such a way as to actually get the sensation across, that I know of, and if I tried with some things in most situations I suspect I’d just come across as crazy. But if I’m talking to someone who assigns the idea that I may experience qualia that they don’t a sufficiently high prior compared to the idea that I am crazy, and I explicitly take steps to bridge the inferential gap (like saying ‘synesthesia’), I can probably communicate that I am experiencing something unusual, even if I can’t communicate what. But this assumes that I have the relevant skills, and a sufficiently sympathetic or open-minded person to communicate with, and it’s fairly easy for me to imagine someone experiencing qualia that are far enough from the things that we have words for that the inferential gap is too wide to cross and they always come across as crazy if they try to talk about it. (We wouldn’t actually notice such cases, I expect, because people in those situations would learn at a fairly young age that talking about the relevant classes of qualia is not useful.)
If we’re just talking about talking itself, though—if you’re simply asking whether I can say some arbitrary word when I experience a particular quale or group thereof—then yes, that’s much simpler. Strictly speaking I’d suspect it’s subject to a degree of false-positive, though. I don’t think it’s all that uncommon to get caught up in a conversation and blurt out something that’s useful, true, and entirely surprising on a conscious level, for example.
Well, as I think about it I realize I’m not really sure what we’re talking about, as the initial context was established by the OP, who was making a point about relationships between talking about experiences and being conscious of those experiences that I’m still not quite sure I understand, and I seem to be keeping one foot in the closest approximation of that context I can manage, which in retrospect is not very helpful of me. I apologize.
So, dropping that context altogether: absolutely agreed that it’s possible for me to have an experience E1 but to not be able to communicate E1 to person X in a way that inspires an imaginative experience E2 in X that we are both confident shares salient properties with E1. And absolutely agreed that when that happens, X may lack confidence that I’m actually having E1 at all… they may think I’m describing E3 in a strange way, or they may doubt that I had any experience at all, etc.
I have a few of those experiences from when I was in the ICU after my stroke that not only can I not readily communicate to others, but don’t even make any sense to me when I recall them.
Come to think of it, that happens all the time, though the most common cases are so conventional that we’ve evolved social standards around them… the experience of holding your child for the first time is one so routine as to be cliche, for example. (There are many other traditional “firsts” in the same vein.) If I try to express those experiences to someone whose life doesn’t include something roughly comparable, I’ll likely fail… but that failure is routine and no big deal. (The traditional wrapper for it is “You’ll understand when you’re older.”)
Agreed that it’s pretty common for me to say things that, prior to my saying them, I wasn’t conscious of knowing, and after I say them, it’s clear I’ve known all along. (This is one reason the “talk about”/”conscious of” equation that started this whole garden path is problematic for me.)
Sounds like we’re on the same page after all, then. ^.^
Upvoted for the possibly-frivolous reason that I’ve never met anyone else whose synaesthesia works this way, and you’re putting words to a basic feature of life experience I can scarcely explain to others.
“Martian colors”
Yep. I’m not colorblind, though—I’m pretty sure the colors I get are just flat-out impossible in the real world.
I wonder if people with blindsight could learn to perceive some of the things in their scotomas. Perhaps if they try to imagine “grabbing the object in their scotoma”, they can feel a sense of feasibility that correlates with there being a graspable object there, without needing to actually grab it. They could learn to constantly poll their mind for this graspability hunch and other hunches, and integrate it into a limited awareness of what their eye is seeing. This would be more useful for people with total blindsight.
I understand all of your sentences, but I can’t tell what overall point they’re trying to make, if any.