About the SA Blog Network



Opinion, arguments & analyses from the editors of Scientific American
Observations HomeAboutContact

Blindsight: Seeing without knowing it

The views expressed are those of the author and are not necessarily those of Scientific American.

Email   PrintPrint

Patient TN and Lawrence WeiskrantzIs it possible to see something without knowing you can see it? Maybe that’s not so hard to imagine if you think of subliminal images flashed for a frame or two on a movie screen—too quickly for you to see consciously but perhaps long enough to add a frisson of fear. (Those frames in The Exorcist don’t count—if you can notice them, they’re not subliminal.)

But what about seeing something when you think you are totally blind? What about navigating around obstacles that you cannot see and aren’t even expecting?

That’s a feat which was performed by a man known to the medical world as patient TN. In 2003, TN lost use of a brain region called the primary visual cortex, which is responsible for processing the visual information that goes on to form our conscious sight. TN’s injury was an extraordinary piece of bad luck, taking two successive strokes to knock out the region in both his left and right hemispheres.

After his strokes, ordinary tests of TN’s sight turned up nothing, not even an ability to detect large objects moving right in front of his perfectly healthy eyes. But researchers had hints that TN might exhibit blindsight, a weird ability to respond to visual information despite having no conscious knowledge of seeing anything. They realized TN’s total cortical blindness presented a rare opportunity, and in 2008 they decided to try out a bold but breathtakingly simple experiment.

They took him to a hallway and asked him to walk along it without his white cane. TN was reluctant, but they finally persuaded him to try. After all, how bad could it be, even being blind, to walk slowly down an empty corridor?


The man right behind TN in the video is Lawrence Weiskrantz of Oxford University. Weiskrantz coined the word "blindsight," and he has been a leading figure in researching the phenomenon for over 40 years—early on, in the face of widespread skepticism. This experiment with TN was inspired by one that Weiskrantz and his then student Nicholas Humphrey carried out in the early 1970s with a monkey with no primary visual cortex. Some of that experiment is also on video.
You can read more about blindsight in the May issue of Scientific American, in the article, "Uncanny Sight in the Blind" by Beatrice de Gelder, who was the lead researcher on the TN navigation experiment. De Gelder has also explored how blindsight can detect the emotion expressed on a person’s face (but, curiously, not a person’s identity or even their gender) and which neural pathways in the brain are responsible for blindsight (it involves the subcortex, which is more primitive than the cortex in evolutionary terms).

And yes, the unconscious effects of subliminal images may be a variant of blindsight, although the jury remains out on just how similar the two phenomena are.

The topic leaves me wondering at random moments just how much my brain is seeing that I don’t know about consciously.

Photo credit: Beatrice de Gelder



Rights & Permissions

Comments 28 Comments

Add Comment
  1. 1. jtdwyer 11:38 am 04/22/2010

    Is it possible that some limited processing of visual signals occurs before it reaches the ‘projector’ that generally presents (highly processed) images to our more flexible abstraction processes?

    While more primitive animals, maybe slugs, for example, may respond to light in controlling their motion, they may not possess the ability to ‘envision’ images for more analytical processing. This low function image processing interfaced to low level motion and perhaps primitive goal seeking processes retained in humans may explain the described phenomenon.

    Link to this
  2. 2. PotatoChip 12:27 pm 04/22/2010

    jtdwyer, I think that this is basically a matter of the reptilian part of the brain that processes visual input being intact, while the mammalian part of the brain that does further processing of visual input being damaged.

    I know that’s a very simple description (and I’m not knowledgeable enough to do any better), but I think it amounts to the same visual competency that reptiles have – in other words, when a reptile "sees" a predator, he sees it, but he doesn’t "know" that he sees it. Same thing with TN.

    Link to this
  3. 3. jtdwyer 12:45 pm 04/22/2010

    PotatoChip – I don’t disagree, except that I think that some animals more simple than reptiles have eyes and some of them are mobile. It make sense to me that there could even be some ‘pre-reptilian’ visual neurons that provide some perhaps direct input to some goal seeking/motion control neurons. Even very primitive vision could be used to detect obstructions.

    Link to this
  4. 4. PotatoChip 1:23 pm 04/22/2010

    True, true. I shouldn’t have implied that primitive visual processing applies to reptiles only. I was just making a distinction between what mammals are "aware" of and what pre-mammals are not aware of.

    Link to this
  5. 5. jtdwyer 4:23 pm 04/22/2010

    Yep, I think we agree. It also seems that the ‘projected’ image that our consciousness is aware of is actually a highly processed and enhanced video stream. As I understand, we actually only detect a small portion of our field of view, but retain other portions and fill in the gaps with guesses to complete the illusion of full vision. In that sense some perhaps specialized memory is involved with vision.

    I’m not sure whether garden slugs can detect motion, but if so they’re apparently not real good at it. Of course when you move so slow, it wouldn’t help much to see that big foot much coming down…

    Link to this
  6. 6. sahar 7:32 pm 04/22/2010

    Consciousness is assumed to be a cognitive ability which perhaps primitive animals aren’t endowed with. In blindsight phenomenon, damage to V1 that is the primary path to higher level areas makes them malfunction. However, it seems beside the retina-V1 connection there are also pathways from retina to extrastriate cortex, not passing V1. Although not sufficient for consciousness, apparently, this information helps blindsight patients respond to stimuli.

    Link to this
  7. 7. jtdwyer 8:21 pm 04/22/2010

    sahar – Thanks very much for your more precise explanation.

    Link to this
  8. 8. whwsjackfrost 9:28 pm 04/22/2010

    So is this case different from the splits where people have had the left and right hemispheres of the brain separated? because if they see a word in there left eye they cant tell you what it is. Or do they know they see it and just cant communicate what it is?

    Link to this
  9. 9. jtdwyer 10:33 pm 04/22/2010

    whwsjackfrost – I apologize in advance if I’m being a little cranky, but I have to ask: can you read?

    The article clearly states that the patient is effectively blind.

    Link to this
  10. 10. Mark Pine 10:47 pm 04/22/2010

    The blindness this experiment demonstrates perhaps resides in brains of the scientists who are interpreting what happened.

    The film clip demonstrates what most people would consider the behavior of a man who is consciously avoiding obstacles that he sees in his path. But apparently the scientists concluded that is not what happened. Instead of hypothesizing magic at work, they hypothesized another unknown "uncanny" phenomenon, blindsight.

    The clip provides the information that when asked, TN stated he had no idea he avoided any obstacles. The most straightforward conclusion, I think, is that the strokes disconnected his language system (which handled the questions from the scientists and answered them) from his visual system (which saw the obstacles and provided information to his motor system).

    Another well-studied example of split consciousness, in which only one part of the division of the brain was able to use language, were the split brain experiments of the 1970s-1980s. The scientist, Roger Sperry, experimented on subjects who corpus callosums had been severed. Experimental probes directed at the right hemisphere showed that it could not receive or respond to language, but Sperry showed that it was able to direct non-verbal behavior. He did not hypothesize that the right hemisphere was not conscious because it did not respond to language. He concluded that the two hemispheres had separate consciousnesses.

    In the film clip, that is what seems to have happened to TN. Consciousness in his visual system seems to have been severed from that in his language system.

    Link to this
  11. 11. jtdwyer 1:28 am 04/23/2010

    Mark Pine – Perhaps they could not confirm the damage to the primary visual cortex on both sides of his brain, suffered in two strokes. Maybe they don’t have PET scans at Oxford and were making the diagnosis based solely on the subject’s explanations.

    Always the skeptic, it also occurred to me that this Weiskrantz of Oxford University, who identified the condition 40 years ago & has been studying ever since, was walking behind the patient so he could whisper which way to go. That’d be a pretty good scam to get over with at Oxford for 40 years…

    Or, maybe they confirmed the damage through other tests and it is as described.

    Link to this
  12. 12. sahar 4:46 am 04/23/2010

    Dear Mark, nice interpretation. Excluding the necessity of having verbal ability, I wonder if TN could be conditioned with visual stimuli. If he does, it could mean the visual pathway to memory has survived intact.

    Link to this
  13. 13. JamesDFuller 8:48 am 04/23/2010

    I remember reading somewhere that some reflexes are triggered directly from the eye, in other words the body responds to some visual cues before the brain even registers. Perhaps this is what drives "blindsight".

    Link to this
  14. 14. Mark Pine 9:12 am 04/23/2010

    In commenting, I was aware that I had not yet read the article. I’m waiting for it to arrive in my next issue of SciAm. I intend to read the article carefully. In particular, I hope the piece shows precisely where the stroke damage occurs and discusses the deficits that might be expected.

    Nevertheless, the film clip makes it very difficult to conclude that the patients visual system was destroyed by the strokes.

    Link to this
  15. 15. Emperor 10:43 am 04/23/2010

    I read an article a few years back about how there at least two visual pathways our brains use. One is the system that we are continuously using every waking moment. jtdwyer described it perfectly as a "video stream." The other one, as I recall, bypasses the visual cortex altogether and feeds directly to the amygdala, the "old brain." This is generally only activated when a danger of some sort is perceived.

    The example I’ll give is anecdotal, but I think most of you will be able to relate. While driving down the highway I generally speed. Sometimes I’ll see a flash of blue light in my rear view mirror, which may or may not be a cop. By the time my visual cortex has realized what I’m "seeing" my body has already tensed up and I can feel my heart pounding. Presumably this is all part of the "fight or flight" response.

    It seems reasonable that when it comes to pure survival a simpler visual pathway that bypasses consciousness altogether would be quicker and more effective.

    "Blindsight" appears to be a manifestation of this.

    Link to this
  16. 16. jtdwyer 12:30 pm 04/23/2010

    Mark Pine – Yeah, just looking at the video I really was suspicious that the patient was getting instructions. Now, if they’d called ‘Dark Vision’, I have been real skeptical…

    Link to this
  17. 17. jack.123 11:17 am 04/25/2010

    I wonder if the people involved have their hearing intact?Is what we are seeing here some form of echolocation.No pun intended.It would be easy enough to use muffs to block sounds to see if this is whats going on?

    Link to this
  18. 18. Mark Pine 5:29 pm 04/26/2010

    Having now read the article, I’m a bit less skeptical, since the authors propose some possible neuroanatomical pathways for blindsight. Nevertheless, my impression from the article is they have a long way to go to demonstrating their claim. I was disappointed that they did show some neuroanatomical diagrams or brains scans of people exhibiting this phenomenon, and that very much lessens the power of the report to convince (although that could possibly have been withheld for reasons of privacy and confidentiality).

    Link to this
  19. 19. Mark Pine 5:31 pm 04/26/2010

    My third sentence should have read: I was disappointed that they did NOT show some neuroanatomical diagrams or brain scans of the lesions of people exhibiting this phenomenon….

    Link to this
  20. 20. mstucchi 7:29 pm 04/27/2010

    that is truly remarkable, i have avoided accidents just in time, without knowing how. i suspect, from some sensory input, outside or below my concious awareness.
    without being to critical, the camera person documenting this, could benefit from this enhanced ability. i was getting sea sick watching their video.

    Link to this
  21. 21. mo98 9:49 pm 04/27/2010

    Blindsight seems to be directly linked to motor coordination. Once perspective processing is automated with the crude peripheral vision of even insect type resolution, the built-in PID (proportional integral derivative) controller of the surviving brain region seems to retain the foundations of learned orientation experience. Much like consciously avoiding obstacles while walking barefoot on the beach while looking 20 steps ahead, and not looking down, it becomes automatic. Now try that with a child…

    Link to this
  22. 22. billvanyo 4:19 am 05/10/2010

    How do we detect consciousness? The person in the video is assumed to have no consciousness of seeing things because they can’t report about seeing things, while on the other hand, they produce some rather complex behaviors that are associated with and indicative of seeing. One complex behavior indicative of seeing – avoiding obstacles, is not considered evidence of consciousness of visual experiences. Another complex behavior indicative of seeing – reporting about what is seen, is considered evidence of consciousness. How do we know which of these behaviors always correlate with consciousness of visual experiences, and which do not?

    Link to this
  23. 23. Harry HB 3:40 pm 05/13/2010

    So, let me play skeptic here: the man behind TN wasn’t giving him cues, either verbally or through his own movements, right or left, slow or fast? His jaw and mouth seemed to be moving in the grainy video, when we could see him (he smiles at the end, after we’ve been watching only the feet of the two men).

    Link to this
  24. 24. jrlamberty 3:17 pm 05/29/2010

    this research is inspiring! I read to a blind student in college and was always amazed her ability! Three cheers for your good work!

    Link to this
  25. 25. Hannes2708 12:57 pm 05/31/2010

    The TN-man who is navigating down the hallway wears obviously a wrist watch, what for, can he blind-sight the watch’s hands too?

    Link to this
  26. 26. Hannes2708 12:58 pm 05/31/2010

    The TN-man who is navigating down the hallway wears obviously a wrist watch, what for, can he blind-sight the watch’s hands too?

    Link to this
  27. 27. jack.123 3:19 am 06/1/2010

    Why the need for the man behind him?I smell fraud,but if this is true a whole new science may be in the offering.Maybe some sort of entanglement of photons is involved.Whole new areas of research may come from this.

    Link to this
  28. 28. chicao teixeira 4:29 pm 07/13/2010

    What about to repeat the whole experience in the same corridor with the same blind guy but at night, with no lights on and the same obstacles in different places? Am I right to suppose that, in this case, the guy would find many troubles to walk?

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American

Email this Article