That Neuroscience Guy - Vision Part 1- The Dorsal Stream

Episode Date: May 2, 2021

Sight is one of our most detailed sensations. In a two-part series, I'm going to discuss the basics of how we process visual information about the world. First, we talk about the neuroscience behind vision for action: determining where things are and how we use them. 

Hi, my name is Olof Kregolsen, and I'm a neuroscientist at the University of Victoria. And in my spare time, I'm that neuroscience guy. Welcome to the podcast. Have you ever wondered about how you see? You look around the world and you just know what everything is and you know where it is. You can just reach out and pick something up easily and you've got this perfect representation of the world around you. Today we're going to talk about the neuroscience of sight and in particular the dorsal visual stream. This is the first in a two-part series in which we'll discuss visual processing in the human brain.

As we all know, visual information enters through the eyes. At the back of the eye, you have receptors, which are called rods and cones, and these are sensitive to photons of light hitting them, quite literally. So photons of light come into the eyes and they hit these neural receptors, and when the receptors are hit, they fire. It's kind of the opposite of how a TV works. In a TV, you've got an image which is made up of pixels, and by turning on those pixels on and off, you can create images on the TV screen. Well, like I said, it's the opposite of that. on the TV screen. Well, like I said, it's the opposite of that. Imagine that the back of your eye is a TV screen, and when a photon hits a pixel, then that pixel is turned on. So you're

building up the image by the photons hitting the TV screen on the back of your eye, which is made up of rods and cones. And when these rods and cones are hit by photons, they fire. From the back of the eye, this pattern of electrical activity, the firing of the neurons, flows through to the midbrain regions. In particular, the pulvinar nucleus, the superior colliculus, and the lateral geniculate nucleus. These are three crucial midbrain nuclei that play a role in orientation. For instance, quickly moving your gaze to orient if something moves by you and helping you pay attention. In fact, there's a really interesting phenomenon called blindsight. You have people that

are cortically blind, as in they can't see, yet they're still sensitive to movement because these midbrain regions aren't damaged. I'll talk about blindsight on a later episode, but if you're looking for something interesting to watch on YouTube, I'd suggest giving it a search. After the electrical signal of the visual information leaves these midbrain nuclei, it makes its way to the back of the brain, otherwise known as the primary visual cortex, or as some neuroscientists call it, area V1. In area V1, you have a representation of the visual world. It's kind of like the TV screen at the back of your eye, but instead it's wrapped around the cortical tissue. In fact, there's some really cool studies where they're able to capture the firing in the

primary visual cortex, and they can literally see an image there in the firing pattern that represents what's actually out in the real world. So the primary visual cortex at its initial layer is just a representation of the visual world that is literally a transform or a copy of what's seen on the back of the eye. In area V1, basically the processing begins, turning that image into something that we understand. And in V1, or the primary visual cortex, there's neurons that are sensitive to patterns, in particular edges and lines, areas of sharp contrast. And V1 is packed with neurons, approximately 140 million of them. And this is how visual processing begins. Area V1 captures a representation of the outside world, then it begins separating what we call low-level features,

the patterns, edges, and lines that I've talked about. Now visual information flows. You can imagine it moving from one region of the brain to the next. And as it moves into the next region, area V2, we begin to have increased complexity of the representation. In particular, there's an attempt to build 3D shape and meaning from what's literally a 2D image. A lot of people find this surprising, but you don't see in three dimensions. You see in two dimensions, at least in the sense that the image that's captured at the back of the eye is two-dimensional, and our brain recreates the 3D world around us. And this is what begins to happen in Area V2. As the visual flow of information moves to area V3, we begin to identify shapes, basic shapes. And this is by combining the edges and lines that were identified in earlier processing stages.

So you might, for instance, have neurons that are sensitive to the fact that there's a square in the environment, or a circle. As you move into the next visual area, V4, you begin to add color to the images and the shapes become more complex. And from there it gets even more complicated. So far there's at least 30 distinct visual areas that have been identified that add separate meaning to our understanding of the world around us. But the real take-home story is it's a gradual buildup of information. The information starts as this very simple representation from which we extract features like edges and lines, we add 3D to it, we increase complexity, we add shapes, we add color, and we eventually get to meaning. But we're not quite there yet. In terms of these other areas, I'll just tell you about one quickly.

Area MT is really interesting. Area MT is a part of your brain that adds motion to the world. What do I mean by that? Well, we don't really see motion, we infer motion. I know that sounds weird, but if you think about an old motion picture where every image was a separate frame, and if you play them quickly one after another, you see motion, because there's a change between each frame. Well, that's what MT does. It is basically helping you integrate subsequent snapshots of the world, and from those snapshots, you infer motion. In fact, there's a very strange condition called motion blindness. People that have damage to area MT, and what you find in these people that have motion blindness is they literally don't see motion. And there's all these

weird examples out there of people having trouble crossing the street because if they turn and look, they see a bus, but they literally don't know if it's moving or not. Now, after these early visual areas, the flow of information splits, and there's a visual stream, if you will. Think of a split in a river where information heads down to the inferior temporal cortex, or what's called the ventral visual stream. We're going to talk more about that on the next episode. Today, we'll focus on the split in the river that heads up to the posterior parietal cortex, or what's called the dorsal stream. Vision for action and vision for spatial processing. So to make sure we're all on the same page, visual information comes in through the eyes, goes through those midbrain nuclei like the superior colliculus that we talked about, eventually arrives in V1, and then that

information is processed and sequentially built up through a series of early1, and then that information is processed and sequentially built up through a series of early stages, and then eventually it splits, and information heads now for the posterior parietal cortex. So I'll tell you a bit about how the PPC works in the dorsal visual stream through a series of experiments. Back in 1987, Dr. Mel Goodale at the University of Western Ontario ran a really cool experiment to begin to probe the workings of the dorsal visual stream. In his study, he had participants reach out to touch targets that appeared on a table in front of them. Imagine sitting in a dark room and all of a sudden a light comes on a table and you just have to reach out and touch it as quickly as you can. Now the trick to the experiment is that sometimes as people were reaching, the target moved.

Now, it moved just a very small distance. And Dr. Goodale and his research team managed to arrange it so that the movement of the target was so subtle that the people in the experiment that were reaching for the target didn't even know that it had moved. the people in the experiment that were reaching for the target didn't even know that it had moved. So they were using a sophisticated motion capture system to accurately measure the position of the arm as people reached to these targets, and they saw something really cool. Even though people weren't aware that the target had moved, they found that people were adjusting their reach to reach the target. So people are reaching out to the target and all of a sudden it moves, but they're not aware it moves because it's a very subtle change in target location.

Yet in spite of this, their dorsal stream kicks in and they're able to accurately reach the target. So that's what the PPC does. It's helping us compute where things are in space. And this is really what it's doing. If you look around your room right now, space. And this is really what it's doing. If you look around your room right now, the dorsal visual stream is your subconscious visual stream. You're not actually aware of what it's doing, but it's actually computing the distance to all of the objects in the room. So it knows how far away the walls are, how far away the couch is, how far away the door is. And it works to build up a representation of this room, but a subconscious representation, a representation that's accurate in terms of measurement, but actually lacks a lot of meaning.

It's just the position of where things are in space. I'll give you one more example. Dr. Goodell also did some interesting work with a patient who's known as RV. Now, RV had damage to the posterior parietal cortex. And one of the experiments they ran with RV was they had a bunch of objects on a table. Imagine a square and a triangle. And they would ask RV, what is this object? And they would show them a square. Imagine a little piece of wood. And RV would say, that's a square. Then they would get RV to reach out and grab the square. And RV wasn't able to do this accurately because RV didn't know where that square was in space. RV could just see a square out in front of them, and they didn't

actually know the position of it. So they would frequently miss. And if they did manage to grab it, they would grab it awkwardly. And this would also be true for the triangle. They would know that the shape in front of them was a triangle, but they wouldn't know exactly where it was. This is known as visual or optic ataxia. It's basically damage to the posterior parietal cortex, and the way it presents clinically is people that have optic ataxia,

typically when they reach for things they miss because they're not sure of where everything is in space. I know this is really, really hard to understand, but these people do see the world around them. They would know that you as their best friend was there. They just wouldn't know where you were. Now, with a lot of these deficits, there's degree of severity. People with severe optic ataxia literally aren't sure of where anything is in space and they can just see things. Again, like I said, it's hard to imagine.

But people with mild optic ataxia just tend to reach for things and miss. So they reach for their glass and they're just off by a bit or they bump into it. Or they bump into objects because they're not exactly sure where they are. People with severe optic ataxia though have a real challenge interacting with the world around them. So let's review the dorsal visual stream. Visual information comes in through the eyes, gets to the back of the brain, the primary visual cortex. It then flows up into the posterior parietal cortex, the dorsal visual stream. And what's happening there

is the dorsal visual stream is building up the world around you. It's placing things in space, but subconsciously. It's the part of the brain that's calculating how far things are away from you and where they fit into the world, but it's just not adding meaning to them. are away from you and where they fit into the world, but it's just not adding meaning to them. On the next episode, we'll add in that meaning and talk about the ventral visual stream. My name is Olof Kregolsen, and I'm that neuroscience guy. I hope you enjoyed today's episode. If you'd like to learn a bit more, you can check out my website on www.olofkregolsen.com, or you can follow me on Twitter at ThatNeroSagGuy. See you on the next episode.