Well, let's consider the physiology now of this part of the brain relative to the inputs derived from two eyes and, and now we move to an animal model system where micro electrodes can be inserted into the brain and the two eyes could be stimulated separately. And one can assess to what degree are action potentials driven by one eyes inputs or the other. So, what we're looking at here is a histogram that represents the copulation of physiological responses recorded from the kitten visual cortex. The kitten was a very common model system for studying the development of vision back in the 60s, 70s and 80s. Now we use a variety of other models including other carnivores such as ferrets and even rodents such as mice and rats are now common model systems for doing this kind of work. And what can be shown at the level of an individual cortical neuron is that that neuron, regardless of what layer it may reside in the visual cortex, is very likely to have a bias in the response of that neuron that will. Tell you something about the weighting of inputs that are driven by the two eyes. For example, if a neuron in the visual cortex is driven exclusively by the contralateral eye, then the experiment will assign it a ocular dominance group of one. If the neuron is driving exclusively by the ipsilateral eye Then the ocular dominance group assignment would be group 7, so this is purely an operational definition, and somewhat subjective, although it can be quantified in, in different ways by different investigators. Well somewhere in the middle then, would be units that are binocular, that is they respond to inputs. Being presented to each eye separately. But they tend to have some bias. So here in the center of this distribution ocular dominance group four would be a totally well balanced set of inputs being driven by the two eyes converging upon the same neuron. So while that does typically yield the greatest number of cells among these 7 different ocular dominance groups obviously there are more neurons that are either biased towards one eye or the other than are perfectly binocular. And so this is fairly typical of your normally developing visual cortex. And so it's not perfectly a normal distribution but it is centered on binocularity, with representation on either side. Now, in kittens, there does, tend to be a bit of a leftward shift in this distribution with, probably a greater percentage of contralaterally biased neurons and ipsilateral biased neurons. So this is what a normal distribution would look like in a typically developing kitten. Now. The experiment that was initially done by David Huble and Torsten Wiesel and their colleagues, involved keeping one eyelid of a kitten closed during this time of typical brain development in early postnatal life. And what they found as a result of this early manipulation of visual experience is a dramatic shift in ocular dominance bias in the visual cortex. So when they repeated this experiment after surgically keeping shut the eyelid of a developing kitten, what they found was that virtually all of the cells that showed visual responses now became responsive to the eye that remained open. Now obviously to do the experiment they had to open the eyelid that had been sutured shut so they can test. Whether that closed eye could now drive visual responses. And the answer is that it could not. Now, what was fascinating about this story is that when they recorded from the thalamus of these kittens, they found that the thalamus was responsive. To the previously closed eye, but not the visual cortex. So this indicated that there was some change at the level of the cerebral cortex that was primarily responsible for, what appeared to be, an aquired blindness of that closed eye. And we call this phenomenon of cortical blindness to the inputs presented to one eye Amblyopia. Now, yet another remarkable aspect of this discovery was that they attempted to repeat the very same experiment in adult animals. And what they found in adult animals Was hardly any change in the relative distribution of, ocular bias and binocularity. And when they repeated the very same experiment, initiated not at birth but at one year of age, what they found was, essentially, a normal distribution. Centered on binocularity. So, so this led to several really important insights. One is that it is possible to change the structure of the circuits that are laid down there in cortical layer 4 as the thalamus. Innervates the cortex based on the balance of activity that are derived from the two eyes. And that the sensitivity of this circuit to this kind of manipulation of experience is restricted, to an early period of post natal life. And this allowed for the definition of what we now commonly call a critical period. So there is a critical period in early life where these circuits are sensitive to changes in activity dependent modulation of ongoing neural activity by sensory experience. Now subsequent investigators went on to ask, what actually is different about the circuitry? What was discovered is, what appears to be a regression of the terminal arborizations of the afference from the lateral geniculo nucleus to the ocular dominance columns that are representing the deprived eye. And commensorately, there is an exuberant growth of the terminal arborizations of those geniculo cortical projections that are being driven by the open eye. So rather than having a balance of ocular dominance columns. The greater expansion of the cortical territory of the afference that are driven by the open eye. accounts for an increase in the size of the ocular dominance columns that serve the open eye at the expense of those columns that serve the closed eye. Which tend to be reduced both quantitatively in terms of numbers of synaptic connections but also in terms of the spacial extent of their terminal arbors. One might have imagined that following the precritical period there is some kind of complimentary and relatively spatially equivalent distribution of afference to the visual cortex that are representing the two eyes. And as the critical period begins to ensue and this circuit becomes especially sensitive to the impact of the experience what one might discover is an expansion of afference that serve the open eye, and a regression of those afference that serve the closed eye, and this is indeed the anatomical picture that was observed. By Hubel Envisel, and Simon Levey, Carla Shats, and, and many others since their time. doing these kinds of experiments in kitten, and in some instances Rhesus Macaque visual cortex. Now you may be wondering about whether these effects can be reversed and Just about all permutation of this sort of experiment that you might imagine has been done. And I won't take the time to talk about the entire cannon of what is known about ocular dominus plasticity in early life. but I will say that it is possible to reverse this effect if We are able to intervene within the critical period. if not, if the critical period passes, an intervention does not begin during that critical period while that window is open. This can result in a lifelong visual deficit. this cortical blindness for the deprived eye may persist across the lifespan. So this raises considerable urgency when treating infants that may have an, a correctable ocular defect in one eye, to prevent this kind of ambliopia for, for setting in. And leading to lifelong visual impairment. Well, for those of you that are interested in vision as I am, I hope you find this a fascinating paradigm to learn about. for those that aren't particularly interested in vision, I want you to know that occular dominance plasticity Has remained a gold standard in the field of developmental neurobiology for understanding the influence of experience on brain development in early life. And continues now some five decades after the seminal discoveries on this topic were first made. this remains the gold standard in the field of developmental neurobilogy. But with any kind of gold standard one does have questions. one might wonder whether this general lesson that we've learned about studying arcing the dominance column development truly applies to all kinds of circuits in the developing brain. Or, are there some somewhat idiosyncratic features of this story that might limit our ability to generalize. So what I'd like to do is to suggest that indeed perhaps the latter is the case. That is we need to explore other kinds of properties of developing cortical networks if we want to understand more broadly the impact of experience in early life. And that what we've learned from studying ocular dominance columns, as powerful as it has been may not inform Us about the developpment of other kinds of circuits.
