So far in this unit, you've seen that the brain represents information in more than one way. You've seen that it uses maps for some kinds of information and meters for other kinds of information. Specifically, it uses maps for visual spacial information and it uses meters to encode the amount of force needed to move the body. This raises the question of how maps and meters interact when we are using vision to guide our body movements. Can maps be translated into meters? Can be meters be translated into maps? It would seem that in order to make an eye movement to the location of a visual stimulus, a map would need to be translated into a meter to actually accomplish the eye movement. Today we're going to talk about how that happens. We're going to talk about a particular brain region called the superior colliculus, which is thought to be a site where information from different sensory systems gets combined and prepared for interacting with the muscles. The superior colliculus is located roughly in the middle of the brain. It's located on top of the brain stem and underneath the overlying cortex. The superior colliculus is involved in controlling saccades and is also thought to be involved in controlling attention. It seems to be a structure that's involved in orienting us in space. And it receives, not only visual information, but also auditory and tactile signals. But today we're going to concentrate on those visual signals. The superior colliculus also contains what is known as a motor map. If you electrically stimulate the superior colliculus, that is if you insert an electrode and deliver pulses of electrical current, you can drive the neurons artificially. This is something that we talked about earlier with regard to cortical area MT and its role in translating visual motion information into commands for smooth pursuit eye movements. When you stimulate in the superior colliculus you can actually trigger an eye movement. That is electrical stimulation in the superior colliculous will actually cause the eyes to move. Now, here's the really cool part where the eyes move depends on where in the superior colliculus you deliver that electrical current. And this was explored by David Robinson in 1972 in a seminal paper. This is a top down view of the superior colliculus. This is the front at this end, the rostral end. This is the back, or caudal end. This is the superior colliculus just on one side. It happens to be the left superior colliculus which we call the left SC. This is the midline over here and this is the side. And what David Robinson found is that when he stimulated at this end of the superior colliculus, he evoked fairly small eye movements. When he stimulated at this end, he evoked much larger eye movements. When he's stimulated closer to the mid line, the eye movements that he evoked had an upward component. When he stimulated closer to the lateral edge, the movements that were evoked had a downward component. So size of movement was organized along the front-back dimension. And the vertical element of the movement was organized in the medial-lateral dimension. The superior colliculus on the other side of the brain has a mirror image map. And I should note that when you make a cecatic eye movement, both eyes move together. So what's happening with the superior colliculus on the left is it is controlling movements of both eyes in the rightward direction. And the superior colliculus on the right is controlling leftward movements of both eyes moving together. So, the superior colliculus has a map of movement amplitudes and directions, but that's not the kind of signal that is ultimately needed to control the muscles. When you move your eyes you use the same muscles for every eye movement. You use horizontally-directed muscles to move horizontally and vertically-oriented muscles to move vertically. But you use the same set of muscles for any movement that has a horizontal or vertical component. You use the muscles in different proportions for these different directions and amplitudes. As you saw last time, the amount of force generated by a muscle, varies in proportion to the number of spikes that the motor neurons impinging upon that muscle deliver. But what I just told you is that the different neurons in the superior colliculus are responsible for controlling different movements. So there are neurons here responsible for controlling those movements and neurons here responsible for controlling those movements. How do signals that are mapped get converted into the kind of meter-like signals needed to contract the muscles in the proper fashion? I'm going to tell you a little about work I've done previously to try to figure out how this might happen. You can read this paper in more detail, if you're interested. The basic idea involves a model of how neurons might convert signals from one kind of format to another. This is now a schematic diagram of neurons potentially in the superior colliculus and the projections they send to a hypothetical neuron located somewhere else. Over here we have the receptive fields that correspond to each of these neurons. So, for example, if there's a visual target located here, this neuron will fire. If there's a visual target located here, this neuron will fire and so forth. Well, what will happen when these neurons fire? Well, a common theory of what occurs next is that a graded pattern of synaptic weights converts this mapped signal into a meter-like signal. So here's how that would work. When there is a stimulus in this neuron's receptive field, the activity would be transmitted to the next neuron via a fairly weak synapse. So it creates a low level of activity at the output. In contrast, when there is a stimulus in neuron B's receptive field, this neuron has a slightly stronger synapse and produces a slightly greater signal in the post-synaptic neuron and so forth. So it's a fairly simple scheme of synaptic weights being used to convert a map signal into a meter-like signal. Well, you might wonder how synapses might have different weights. What does that even mean? Well, this could involve having more neurotransmitter released at some synapses than at others. Or it could mean having more receptors on for a given neurotransmitter on the post-synaptic neurons. There are any number of ways that the strength of a synapse can vary. This particular model has a couple of problems and they relate to the question of how the brain might accomplish normalization when there are different numbers of stimuli or when stimuli vary in factors like brightness or when electrical stimulation is being used to manipulate the activity in the map. We saw in an earlier video concerning stimulation that some of these same issues arise when considering how the map of visual motion in area MT maybe read out to produce smooth pursuit eye movements. That's a little bit more in depth than we have time to go to in this class, but if you're interested in these issues, I encourage you to go read the paper. In the next video I will return to the question I started with a few videos ago, mainly how is sound location encoded. Now that you know all about maps as well as meters, we're ready to discuss what happens with sound.