Well, I tell you all this so that I can describe for you some experiments that my colleagues and I have done to look at how the experience of the animal in early life impacts the self-organization of these cortical networks that perform these computations that account for orientation selectivity. And direction selectivity. And we began by simply looking at the emergence of these orientation preference maps in the visual cortex of animals beginning from just before the time they naturally opened their eyes, and then on through the phase of normal cortical development. And so if one looks at an animal soon after it has open its eyes, one can indeed find that their is a system of orientation columns that are organized into a map of orientation preference. So, this has been known since really, the early 90s, when it was first possible to apply these methods to the developing visual cortex. And, what was discovered is that these columns do indeed self-organize. They appear pretty early in postnatal life, and those animals that are born with their eyes closed, they seem to emerge prior to the time of eye opening but they rapidly develop in the days to weeks that follow the onset of normal visual experience. So we thought an obvious experiment to test whether vision had an important role to play in the maturation of this system, was to raise animals in total darkness. And, thus removing visual experience from the life history of the animals. And when we did that experiment, what we found was that sure enough. The circuits for orientation selectivity and preference do self-organize. It's possible to recognize. It's possible to recognize columns and the response maps, and in the selectivity maps that we generate from these animals. And the signals are sufficiently robust. That we can assign color values and create an orientation preference map. With the proper analysis applied to these maps, indeed, these maps from animals that never saw any photons of light until the experiment, do indeed self organize and develop. Pinwheels with a density of Pi. However, I think, perhaps, even your eye can pick this out. the contrast of the images that we recorded from these animals that lacked vision seemed to be sig, significantly less than what we observed from animals that had normal vision. Throughout the period of brain maturation that we study. So, while these cortical networks do indeed self-organize without the benefit of vision, vision does seem to contribute to the refinement of these maps, and an increase in the selective responses. Of the neurons, whose signals we're recording with this method. Well, just because these circuits self-organize, and are present in animals that have never experienced vision does not mean that this process. Cannot be influenced by experiment. So we thought it was important to do a complementary experiment. Rather than just taking away vision we thought we should make vision abnormal. And a fairly straightforward way to do that is to raise animals keeping their eyelids shut. And that's very much like what your eye might do if we were to close our eyes and then look around. The visual scene, if I look up towards the florecent lights in the ceiling I can tell that I'm looking up. And seeing an incease in illumination compared to if I were to look away. But I see no form. I see no structure. I see no shape to what. Eye, eye view, simply an increase or a decrement in light intensity. So this is the nature of visual experience that we imposed on the set of animals, and when we did this, what we discovered is that there were some pretty significant impacts on the development of the circuits that compute orientation preference. Now this manipulation did not produce cortical blindness. And we know that because when we looked at the response maps in the visual cortex, we saw robust activation of the visual cortex. The problem however, is that we failed to see the robust development of columnar structure It's as if the very same circuits in the brain responded to the presentation of a horizontal stimulus and the presentation of a vertical stimulus. So when these images were subtracted, what we found was virtually no evidence of columnar structure that was organized into a map of orientation preference. So, I want to notice the implication of, this finding here. What this shows us is that when we made vision abnormal, what we found was actually more significant imparement of the visual cortex then when we simply deprived the animal of vision. Now, this reminds me, at least, of a famous, aphorism. That was suggested by, one of our, seminal neurosurgeons of the 20th century. Dr. Wilder Penfield. Who said that no brain is better than bad brain. Well, that justified some neurosurgical procedures aimed at the removal of bad brain. In this case, I would suggest that bad experience is worse. More detrimental to the development of visual cortical circuits, than is no experience. So what have we learned from these studies? I think we can make some provisional conclusions based on these experiments looking at the development of orientation preference.`` I think what we've learned is that normally circuits in the visual cortex self-organize. And they operate synergistically with normal sensorimotor experience. And this synergy promotes the full maturation of these circuits. However, when experience is rendered to be abnormal, then this synergy is broken. And self-organization goes awry. [INAUDIBLE]. The neural circuits that develop as a consequence are functionally impaired. Now, these neural circuits, they self-organize to adapt to the quality of the incoming sensory signals. So, by self-organization going awry, I mean relative to the synergy that our brain has evolved to anticipate From the world in which we live. But when that synergy is aggregated by some kind of ocular impediment, in our case, keeping the eyelids closed or perhaps in clinical populations congenital cataract, for example. would be one human condition that we would model through this method of keeping the eyelids shut. well, such conditions. Keeping the eyelids shut or congenital cataract. Are those that, would provide for the adaptive influence over the circuity of the visual cortex that would then develop. So consequently, not only do these neural circuits under these conditions fail to benefit from normal visual experience. And as a result, they're actually, developing along a trajectory that instantiates functional impairment. So now I'd like to turn our attention to the property of direction selectivity and direction preference. So we can think of this as a further differentiation of the circuitry for orientation selectivity and orientation preference Where those circuits begin to specialize for the representation of motion in just one direction or other, with the preferred direction of motion being orthogonal to the preferred orientation. That is to say, circuits might develop a preference, let's say, for vertical stimulus. but there might be a differentiation of one side of the orientation domain to prefer vertical moving to the right. Whereas, the other side of that orientation column might differentiate preference for vertical moving to the left. Now we can resolve this with the same method I have just described for you this optical imaging of intrinsic signal methods. One can even resolve it to the cellular level using a method known as two photon imaging of calcium signals which is what here is represented here to the right. So it's now possible to do studies where, for example, we can, zoom in on a small region of the visual cortex and recognize that, with this more low-resolution method, we can define a region that responds, let's say, mainly to downward motion And that region representing downward motion seems to be separated by some kind of barrier condition from an adjacent region that represents upward motion. We imagine that within this hyper column here, there is an orientation column that represents horizontal. But it has subdivided into these two adjacent domains that represent opposite directions of motion. And then, zooming into this region with this method that provides cellular resolution, we see, indeed, there is a population of cells, each cell being represented by an arrow here that has a preference. And the upper right region of this field seems to show a predominant preference for upward motion, whereas the lower left region of the field shows a predominance of cells that prefer downward motion.
