Episode Transcript
[00:00:04] Speaker A: Id the Future, a podcast about evolution and intelligent design.
Welcome to id the future. I'm your host, Andrew McDermott. Well, today I'm joined again by physicist doctor Brian Miller to continue our discussion of the intelligent design of human vision. Doctor Miller is a senior fellow of Discovery Institute's center for Science and Culture, where he serves as research coordinator. He holds a b's in physics with a minor in engineering from MIT and a PhD in physics from Duke University.
He helps to manage the CSCs ID 3.0 research program, and he speaks internationally on the topics of intelligent design and the impact of worldviews on society. Brian, good to have you back.
[00:00:50] Speaker B: It's a pleasure to have you back.
[00:00:51] Speaker A: Well, in part one of this conversation, you discussed the evolutionary scenario for how the vertebrate eye came about. We talked about the fact that even Charles Darwin thought it absurd in the highest degree, that an unguided process like natural selection and random mutation could account for the human eye. You also explained to us why it's helpful to approach biological systems from an engineering standpoint. And you reported on some of the work being done by the engineering research group that you're working with. At the end of part one, you walked us through some of the subsystems and processes at work in human vision using, you know, how we go from light gathering to a high definition image that we can comprehend?
Maybe, and you don't have to repeat yourself fully on that question, but maybe you can start here by just briefly summarizing how our brains comprehend the image that we're seeing with our eye and how it processes that. Just, just briefly.
[00:01:46] Speaker B: Oh, certainly. So the way that works is you have the cornea and the lens, which focuses the light, so it produces a high resolution image on the retina. The retina is essentially an array of photodetectors, like 100 million photodetectors, which then digitizes the image. And then what happens is the image is preprocessed through a series of neural networks in the retina, and then the image is sent through the optic nerve to the brain, and then the brain will process the image and reconstruct it, and then pass the image onto higher level systems to know how to respond to it. You also have feedback loops where the pupil will get larger or smaller depending upon the amount of light. You have specialized muscles and feedback loops which change the dimensions of the lens to make sure the light is always focused. And then you have other feedback loops that causes your eye to move in the right direction at the right time to ensure that you're always focused on a particular image so it doesn't look blurry. So that's just sort of a quick overview of many of the systems related to vision.
[00:02:48] Speaker A: And what I found fascinating, too, as I was studying this is all this data gathering and signal and image processing is done almost instantaneously. I mean, we're talking measured in picoseconds, where a picosecond is 1,000,000,000,000th of a second. I mean, how fast is that?
[00:03:04] Speaker B: Yeah, you're right. It's incredibly fast. And what happens is everything in vision is highly optimized. So, for instance, if you look at the photoreceptor, it'll detect photons and process the signal very, very fast. But what's really interesting is there are physical limits of how fast the image, or the neural impulses can go from the eye to the brain. So this is what's truly amazing, is what happens in the retina, is there's a neural network that anticipates the time it takes for the image to go from the retina to the brain. And then what it does is it actually will send an image a little bit in the future. So let's imagine someone throws a baseball. What happens is, by the time the image of the baseball reached your brain, the baseball might travel several feet. So the image that goes to your brain anticipates the future and shows the ball where it actually is when the image is actually processed. So it's extraordinary how the different time scales are linked together to ensure that you always have a continuous, real time vision image.
[00:04:09] Speaker A: Yeah, that's just fascinating. Well, at one point in your talk that you are preparing and presenting to people on this, you do mention dgrns, developmental gene regulatory networks. Can you remind us what those are and how engineering principles like control theory might help us understand the role of dgrns in human vision?
[00:04:30] Speaker B: Oh, yes, certainly. What happens is a developmental gene regulatory network is simply several genes that work together to control the process of development. And it essentially is a control system. And what happens is one gene will activate another gene, which activates another gene, which then will cause the right developmental system to initiate at the right time, in the right place in an embryo. And vision is pretty, quite remarkable, because what happens, you have all the cells that will eventually become the parts of your eye, your cornea, your retina, your photoreceptors, your lens, a muscle tissue. And what happens is these cells, during development, send signals to each other, and those signals will then activate different parts of your developmental gene regulatory network. That causes the cell to do the right thing at the right time. And the precision necessary is extraordinary. Let me just give you one example. If you look at a lens, it seems like a very simple structure. Essentially, it helps to focus light, but its manufacturing process is incredibly complex. So what happens is a cell actually will elongate and form a fiber cell. And then what happens is at the right time, it will eject its nucleus and other organelles, and then it has to fill up almost entirely with the crystalline proteins, which allow it to be transparent. So all this has to happen in perfect timing, because if you, let's say, eject your machinery too soon, you can't fill up with the crystalline and you'll have an opaque object that'll block your vision. Also, what has to happen is special channels have to form in the lens so that you have this aqueous humor that can then deliver the right nutrients to the lens because there's no blood vessels. So again, what you find is what seems very simple is an stunningly complex system that has to be perfectly timed to properly manufacture the lens and every other component, and it's all controlled by these DGRNs. And what's key is this is where evolutionary theory completely collapses, because again, evolutionary theory says everything has to happen incrementally, in small steps. But if you don't have a lens, the first step to have a lens is to allocate tissue. But that tissue will actually block vision and be deleterious to the organism until the entire network is in place to properly manufacture the lens filled with the crystal material so it's transparent. So you have to go from a system without a lens to a system with an entirely manufactured lens in an instant, or else the organism will not be able to outcompete other organisms.
[00:07:17] Speaker A: And this is where the irreducible complexity of such a system comes into play. Is that right?
[00:07:22] Speaker B: That's absolutely the case. In fact, when you talk about irreducible complexity, the level of irreducible complexity is just staggering, because you're not just dealing with an irreducible complex system. What you're dealing with is a system of irreducibly complex subsystems where every subsystem has to be in place. You've got to have manufacturing, you've got to have the ability to change the size of your pupil, you've got to be able to detect photons, you've got to be able to process the image, you've got to be able to pre process it, you've got to be able to maintain the systems, which is not easy. So you have multiple systems that have to be in place at once. And every system has subsystems, and every subsystem will have sub components, which are all irritably complex. So the level of irreducible complexity and the tightness of the constraints is absolutely staggering.
[00:08:16] Speaker A: Wow. Well, I know the engineering research group wants to expand on the initial analysis done by Laufman and Glixman in your design body. What topics are they addressing?
[00:08:28] Speaker B: Well, we are at the earliest stages, so what we're doing is we're categorizing different possible topics of investigation and then trying to decide where does the skill set of our participants match, where there's a lot of biological literature on the topic, which matches a place where engineering principles can be easily applied. So let me give you a few examples. I mentioned this issue of irreducible complexity that's hierarchical. And you really see that with something as simple as giving nutrients to your lens and your cornea, because that's just one subsystem, which is the maintenance of the lens in the cornea. But that subsystem has its own subsystems. So, for instance, because you don't have blood vessels around your lens and your cornea, you have to produce what's called an aqueous humor. And that's a liquid that resides in that chamber, and it provides all of the glucose, all the nutrients, and it also is able to remove the waste. So you have to constantly produce aqueous humor with the right materials, and there's multiple feedback systems to allow that. But you also have a canal which removes liquid, and that allows you to remove the waste. And what happens is the pressure has to be maintained within tight constraints, because if you have too much pressure, then you're going to damage your system. If you have too little pressure, you won't be able to support the structures in the eye. So there's very complex pressure sensors, there's feedback loops, there's systems to deliver more aqueous humor, which is a beautiful system that you can study from the principle of control systems. Another example is image processing, and that's both in the retina and in the brain. And we have a professor who's an expert in machine vision, and he's working with our biologists to figure out how is it that humans, or how do we design our computers to process images, let's say, for an uber car that's self driving? And how do those principles translate to human vision? What's similar, what's different, and what's truly amazing is that you really see the same design logic in human vision as you see in machine vision, except the human vision is far, far superior in many, many different ways. Another example would be eye movement, because if you imagine you create a robot and that robot has machine vision, then it has to be able to process the vision in the right way. And what takes place in the brain is we've got these semicircular canals, which I alluded to, and they basically will detect motion if we're moving our head left, right, up, down, if we're jumping, if we're running. And what happens is that sends feedback to our visual system, so your eyes will keep focused on the same point, so that way we don't have blurry vision. So this issue of eye movement connected to acceleration sensors is a beautiful thing to study. And, of course, manufacturing. I already talked about where you have to deal with delivering the right materials at the right time. You've got to have schedules for each stage of manufacturing. And that's sort of another example of something that we can study. And there's many more examples. So there really is incredible richness in what we can study. We just have to figure out what will be the most efficient and effective for us to start with.
[00:11:50] Speaker A: Yeah, and lots of questions and lots of additional avenues for research on offer there. Well, in recent years, there have been some claims that parts of the human eye were designed badly, an effort to discount the possibility of intelligent design. But the closer you look at these systems, the more engineering prowess is evident. How does the eye represent optimal design?
[00:12:13] Speaker B: Oh, that's a wonderful question. What happened? And this is a pattern you see over and over again in evolutionary literature. I call it the imperfection of the gaps argument. Because what happens is when biologists look at a system, if there's something that doesn't make perfect sense to them immediately, they'll say it's probably because it's poorly designed. It's just the fact that we evolved through this undirected process. So a lot of what we have is probably clumsy and inefficient. That's when the assumption, and time and time again, what people initially thought was poorly designed was later shown to be optimally designed. And that would be everything from the appendix, everything from certain nerves in our body that take long paths, to countless organs that biologists originally assumed were nonfunctional. But now we know they're optimally designed for purpose. And one of the best examples of that with human vision is the fact that our photoreceptors don't face forward, but they face backward. So, like, if you look at a cephalopod eye, like an octopus eye, what happens is the photoreceptors face towards the light, and then the signal goes to neurons in the back of your retina, an octopus's retina, which then goes into their brain. But in human vision, what happens is it's backwards in with respect to an octopus eye, because the photoreceptors face backwards, so the light has to actually go through the neural tissues before it gets to the photoreceptor. And biologists said very often that that is a terrible design, that if it was created by a designer, no way would a designer do that. But what's happened is, as our understanding of vision has increased, our understanding of the eye has improved. We realize that that is not a poor design, but that is an essential design feature, because unlike an octopus photoreceptor, our photoreceptors use much, much more power because we're in the brighter light. So what happens is the photoreceptors are composed of layers of discs, and that's where the rhodopsin proteins are housed. What happens is those discs are constantly burning out. They're damaged because of the radiation, which are photons. So what takes place is if our photoreceptors face forward, the photoreceptor would quickly burn out and be useless, and we would be blind. So what happens is they face backwards. So they're embedded in the retinal epithelium tissue. And what takes place is there's a very special mechanisms where the retinal tissue works with the photoreceptor to essentially peel off the burned out discs and then recycle the materials that were in the. That the disc contained. So that way, the photoreceptors are constantly renewed, which is why they last for our entire lives. And again, imagine creating a photorespect, something like a photo optic plate for a human device that would last 80 years. We would not be able to do it. It would have to be replaced long before that. So it's pretty amazing. In addition, what happens is the photoreceptors require enormous amounts of fuel and the replacement of certain molecules. So the special interactions with the epithelium tissue allows for all that fuel in those replaced materials to efficiently be delivered to the photoreceptor. If the photoreceptor faced forward again, they couldn't get the nutrients it needs to be able to work at an optimal level. So, again, that's just an example of how people thought something was poorly designed. We now know it's optimally designed. And in particular, another amazing discovery is there's very specialized cells that act like photo optic cables. So light will actually go to the retina, and then these cells will deliver the light right to the photoreceptors. And I believe they're even optimized for different colors, depending upon the cone they attach, they connect to. And what happens is, these specialized cells have fibers that actually employ quantum confinement to efficiently deliver the light. So you're seeing quantum nanotechnology at work, which is actually pretty amazing. But beyond that, what happens is photoreceptors have a remarkable ability called perfect, robust adaptation. So what takes place is a photoreceptor can detect a single photon and deliver that image to your brain if it's very dim light. But if the light is, let's say, a trillion times more bright, we can also see in that very bright light, because what happens is there's multiple feedback systems that change the sensitivity of the photoreceptors and change the rate at which signals are sent to the brain such that it constantly adapts to allow us to see light with remarkable differences in intensity over a factor of, let's say, a trillion. And in addition to that, the efficiency by which photoreceptors detect the light and send signals has been estimated to be very close to what's maximally possible based on the physics and chemistry. And there's work out at Princeton, I believe it was Baelik that has shown that it is optionally designed, even based on the physics. So what happens? The more we study vision, the more astounding it is, both in terms of its optimality and its ingenuity for detecting images and then processing them.
[00:17:32] Speaker A: Yeah, I mean, that vision is possible from the get go is a startling thing, and not something you would expect, um, right away, at least, especially given on the. Given the evolutionary perspective, you wouldn't expect such a complex, interconnected system that then actually works, you know? And that is a point that Laughman and Glixman make in your design body. They say vision requires more solutions to more difficult problems than perhaps any other system in the body. It combines perfectly tuned biochemistry with solutions to complicated engineering problems involving general physics, optics, and electrical engineering. As you've been studying this system, what is your favorite aspect of it that just blows you away?
[00:18:19] Speaker B: There's so many amazing aspects to it that it's impossible to say which is the most astonishing. But one aspect I think I'll say was most surprising was the difficulty of maintenance, because it is exceptionally difficult to ensure that photoreceptors and the neurons in the retina can be maintained. And that requires incredibly sophisticated feedback systems to ensure the right materials are delivered to the right place at the right time. Also, what's amazing is, in the same way there's a blood brain barrier, there is a retinal blood barrier, too. And what happens is what the retina does is there's special protective tissue that ensures that things like your immune system cells don't go in and start damaging these very delicate neural connections. So there's a very different sort of immune system at work inside your retina that doesn't damage the tissue, so you don't have things like inflammation, which is an additional irreducibly complex system. So I think the extent to which the retina only allows the right molecules in and the right molecules out, in my perspective, is incredibly stunning.
[00:19:33] Speaker A: Well, I am looking forward to what you and others in the engineering research group are going to work on in this particular area, human vision, the vertebrate eye. And you are welcome to come back anytime and update us on this most amazing system.
[00:19:49] Speaker B: Thank you. I look forward to talking more about this topic.
[00:19:51] Speaker A: Well, thank you, Brian, for coming on the show. And as we said, we'll link in the show notes for this episode, some of Brian's previous writing on the vertebrate eye and refutations of the evolutionary standard that people put forward for this. So you'll find that, as well as information on how you can get your hands on a copy of your design body by systems engineer Steve Laufman and physician Howard Glixman. For now, I'm Andrew McDermott with Brian Miller, and we're happy to share this information with you for id the future. Thanks for listening.
