To See, but not to See

Posted by megkirch on September 30, 2016 in neuropsychology, Perception, vision | Leave a comment

In 2008, a man shocked researchers when he successfully walked through a hallway cluttered with boxes, trash cans, and other junk without stumbling into anything. That may not seem terribly impressive (particularly if you’re used to navigating through a messy bedroom every night), but here’s the catch: that man was blind.

That man, known as patient T.N., is different from other blind individuals you may know or have heard about; his visual system developed completely normally, and his eyes are totally intact. Instead, T.N. has cortical blindness, meaning that his loss of sight resulted from damage to his primary visual cortex (aka “V1”). Think of V1 as the quarterback of the visual system. Once it gets the visual information from the eye, how, when, and where it decides to “throw” that information determines the whole rest of the play, i.e. your visual experience. And like the quarterback of a football team, V1 is thought to be necessary for normal vision. In fact, V1 was first identified and named in the late 1800’s because humans and animals who sustained damage to that part of the brain became blind, like T.N. [1].

However, he and other patients like him demand a reevaluation of V1’s role in vision and if it is, in fact, necessary. They are still able to perform tasks and behave in ways that suggest their vision is at least partially intact, even though they will insist that they can’t see anything at all. This phenomenon, provocatively and oxymoronically named “blindsight”, is a fascinating example of just how complicated the brain really is and how vision, like football, requires more than just a quarterback.

 

Seeing without seeing: The blindsight phenomenon

If I were to show you two shapes and ask you to identify which of the two shapes was a square, you would (presumably) have no trouble identifying the correct shape. If you were completely blind, however, you would have to blindly (pun intended) guess which was the square and would get it right about 50% of the time. Psychologists Elizabeth Warrington and Lawrence Weiskrantz gave this task to another famous patient known as D.B. who had sustained damage to V1 on the right side of his brain, leaving him completely blind to half of his visual field. Remarkably, when D.B. performed this task with the shapes presented to his blind side, he successfully identified the correct shape almost 90% of the time despite reporting that he hadn’t seen the shapes at all [2]. In another series of experiments, D.B. was even able to correctly identify and name 25 out of 28 low-contrast outlines of animals, even though he insisted he couldn’t see anything [3].

Although D.B. and T.N. are a few of the most famous patients with blindsight, theirs are not isolated cases. This phenomenon has been demonstrated in both humans and monkeys with V1 damage [4]. It’s also evident in many other types of experiments; for instance, subjects with cortical blindness still blink in response to a flash of bright light and can orient their eyes toward a visual stimulus with remarkable accuracy (reviewed in [3]).

 

Understanding blindsight: Different routes for vision

How is it possible that subjects with damage to the part of the brain that is supposedly essential for vision are able to “see” without being conscious of it? In order to grasp how blindsight might occur, it’s necessary to understand how typical vision arises in the brain. Imagine vision like a package of items you ordered online. First, neurons in the thalamus – the brain’s sensory warehouse – have to receive visual information (your online order) from neurons in the retina. The thalamus also receives auditory information from your ears, sensory information from your limbs, etc., but the thalamus is organized according to the type of sensory information being received and transferred. The part of the warehouse that handles visual information, called the lateral geniculate nucleus (LGN), relays visual information to V1 – the visual distribution center – which packages that information and then sends it along to other “higher visual” areas of cortex for further processing and shipping (Fig 1, blue pathway). In real life, having too many stops along a shipping route would not be the most efficient means of package distribution. However, in the visual system, a lot of important stuff happens at each station to change the package and get it closer and closer to the finalized product. For instance, while neurons in V1 are sensitive to very basic visual features like oriented lines, neurons in the higher visual areas are generally believed to be sensitive to increasingly complex visual features, such as the direction a particular visual stimulus is moving. These increasingly sophisticated features are critical for your final product of vision so that you can see a car making a quick lane change in front of you on the freeway or a shooting star moving across the sky. Moreover, pretty much every station in the visual system gets feedback from each station it sends information to. It wouldn’t make much sense for a package to go backwards along its shipping route, but these feedback connections are pervasive throughout the visual system and are hypothesized to be particularly important for visual awareness [5].

In this hypothetical shipping scenario, there are a number of possible routes that a package could take through the various stations before finally being delivered. Similarly, this retina → LGN → V1 → higher visual areas pathway is not the only path that visual information can take through the brain. For instance, some online orders can get processed at and shipped from another distribution hub, the superior colliculus (SC). This is possible because the majority of neurons in the retina directly connect not only to neurons in LGN but also to neurons in the SC [6]. The SC can then pass its information along to LGN or, to make things even more complicated, to a different part of the thalamus called the pulvinar. Moreover, both LGN and pulvinar are able to bypass V1 and send information directly to shipping stations in the higher visual areas [7]. Which shipping stations your package stops at could depend on the source of the package, whether you specified Standard or Two-day shipping, or countless other possible variables. Moreover, even though V1 is the main distribution center for vision packages, there are instances in which your package, whether randomly or by necessity, gets redirected to one of the other possible stations.

Once we understand that there isn’t just one possible pathway that visual information can take through the brain, blindsight starts to seem a little more plausible. Nevertheless, if the V1 distribution center is shut down, there isn’t going to be a seamless redirection of all of the packages that would normally pass through. V1 was more than just a holding station; it did some important repackaging, binding together different products that you ordered at the same time, and relabeling to facilitate transfers to the higher visual stations. So even though some packages will make it all the way to their final destination, they might get damaged along the way, missing one (or more) of the products you ordered, or just be very late. Thus, you can think of blindsight as an imperfect package of vision that still manages to influence your behavior through one of the other many possible shipping routes, even when V1 is out of commission.

But are some of these visual shipping stations more important for the blindsight phenomenon than others? For a long time, it was believed that the retina → SC → pulvinar → higher visual pathway (Fig 1, black pathway), which completely bypasses both LGN and V1, is what enables seeing without seeing. This hypothesis mainly came from findings that the SC is necessary for visually-evoked neural activity in the higher visual areas in the absence of V1 [8]. However, a few years ago, researchers discovered that the LGN is also critical in times of shipping crises. When monkeys’ V1 was damaged but their LGN was intact, they correctly detected a visual stimulus, and activity in their higher visual areas also changed in response to the visual stimulus. However, both of these effects disappeared when the LGN was temporarily shut down, strongly suggesting that LGN is necessary for blindsight [9]. Still, the SC is likely very important for blindsight as well. Given that the parts of LGN that are known to pass information along to higher visual areas are the same parts that receive information from SC [10], the retina → SC → LGN → higher visual areas pathway (Fig 1, red pathway) may be the critical pathway for blindsight.

 

Conscious vs. subconscious vision

Even if humans and monkeys are still able to demonstrate behaviorally that they can effectively “see” without V1, doesn’t it say a lot that their conscious experience of seeing is lost without V1? Interestingly, it is still hotly debated whether V1 is actually even needed for visual awareness [3,7]. For instance, some patients with V1 lesions who demonstrate a slightly different sort of blindsight are conscious of the presence of a visual stimulus without literally being able to see it [3].  Moreover, in healthy humans and monkeys, V1 activity in response to a visual stimulus seems to be independent of whether the subjects consciously perceive it. This has been shown using an experimental paradigm called perceptual suppression in which the subsequent presentation of lots of moving dots sometimes, but not always, makes an initial target stimulus invisible for a few seconds (Fig 2). When comparing the activity of individual neurons on trials when the target was visible to trials when it was invisible (but still present), neuronal activity does not change [7]. This raises a lot of fascinating questions about how and where the experience of conscious perception arises in the brain that have yet to be answered.

So, do you actually need V1 – the “quarterback of the visual system” – to be able to see?  It all depends on how you define “seeing”. Patients with V1 damage like T.N. and D.B. lack conscious vision, but their behavior and performance in experimental tasks suggest that some form of subconscious vision remains. Overall, the blindsight phenomenon demonstrates that our cumulative visual experience is the product of many brain areas, not only V1 but also the LGN and pulvinar of the thalamus, the SC, and the higher visual areas, all working together to allow us to perceive and respond to the world around us. When V1 is gone, our experience of the visual world may disintegrate, but a little bit of the visual world can still, remarkably, find its way into our brains and influence our behavior. Vision is a team sport, and you need more than just a quarterback to win the game.

 

References:

[1] Tong, F. (2003). Primary visual cortex and visual awareness. Nature reviews. Neuroscience, 4(3), 219–29. doi:10.1038/nrn1055

[2] Weiskrantz L (1987). Residual vision in a scotoma: follow-up study of form discrimination. Brain, 110, 77–92

[4] Stoerig, P. & Cowey, A. (1997). Blindsight in man and monkey. Brain, 120, 535–559.

[5] Lamme, V.A. (2001). Blindsight: the role of feedforward and feedback corticocortical connections. Acta Psychol., 107, 209–28. http://invibe.net/biblio_database_dyva/woda/data/att/4fc3.file.pdf

[6] Ellis, E. M., Gauvain, G., Sivyer, B., & Murphy, G. J. (2016). Shared and distinct retinal input to the mouse superior colliculus and dorsal lateral geniculate nucleus. Journal of neurophysiology, 116(2), 602–10. doi:10.1152/jn.00227.2016

[7] Leopold, D. A. (2012). Primary visual cortex: awareness and blindsight. Annual review of neuroscience, 35, 91–109. doi:10.1146/annurev-neuro-062111-150356

[8] Rodman, H. R., Gross, C. G., & Albright, T. D. (􏱾1990). Afferent basis of visual response properties in area MT of the macaque. II. Effects of superior colliculus removal. Journal of Neuroscience, 10, 1154-1164.

[9] Schmid, M. C., Mrowka, S. W., Turchi, J., Saunders, R. C., Wilke, M., Peters, A. J., … Leopold, D. A. (2010). Blindsight depends on the lateral geniculate nucleus. Nature, 466(7304), 373–7. doi:10.1038/nature09179

[10] Harting, J. K., Huerta, M. F., Hashikawa, T. & van Lieshout, D. P. Projection of the mammalian superior colliculus upon the dorsal lateral geniculate nucleus: organization of tectogeniculate pathways in nineteen species. J. Comp. Neurol. 304, 275–306 (1991).

 

Images:

Title image: José María Pérez Nuñez

(https://www.flickr.com/photos/jmpznz/5081826831/in/photolist-8K4G4Z-pKTS3T-pL8pwT-ptG4cT-pKTKiF-pLcLtw-9zTok3-e8Zm7H-74LmJn-CMowCg-mwgcc2-9oypFu-4peGDK-3etbei-3ethzd-86MmJK-3eqg6x-3eq5eD-aYmfnV-mwhJCs-5T4eGK-3etgDh-5L3nYf-3etfYJ-9oSkWC-3eoTBx-8ve3Tz-3epZzr-4peG9R-4peGoR-mwgnug-3euxbE-4peEf8-4uodNE-5M4TKV-4peMqn-8zuBho-aw1qjx-98PmCX-4peBKn-4pisT9-4pepxv-3eqZvB-3euN1C-3ewbbW-8hQK47-4piodG-4pez1p-3eqvNg-4fzEUc)

Blindsight cartoon: https://neuwritesd.files.wordpress.com/2016/09/0790e-image2b35.jpg

Vision brain: modified from https://upload.wikimedia.org/wikipedia/commons/2/20/Brain_circuits_for_visually_guided_saccades.jpg; attribution: By Robert H. Wurtz [CC BY 4.0 (http://creativecommons.org/licenses/by/4.0)%5D, via Wikimedia Commons

Perceptual suppression: modified from [7]