Professor Ohzawa’s academic background is in electronics engineering, but he has extended his expertise into neuroscience as well. As he sees it, electronic engineering and neuroscience are not so dissimilar. This is because the approach to understanding how the brain sees things is in some ways similar to the image processing technologies used in televisions and cameras.
Quantifying what one brain cell sees
In my research, I aim to develop a quantitative representation of what a single brain cell sees when a visual stimulus from the outside world is received by the eye and transmitted to the brain. Individual cells only see a tiny fraction of the whole scene, but this is not in a form like a piece of a photograph that has been cut up. The picture is divided up in a much more complex manner. If we are able to explain this division to the point that it can be reconstructed on a computer, there will be great scope for practical applications.
Starting at the primary visual cortex with a view to higher-order functions
In my lab, we expose animals to various visual stimuli and use computers to analyze the response of individual brain cells when those stimuli are received. Research to date has enabled us to know roughly how cells will react to certain patterns. In our experiments we often use a striped pattern. Different cells prefer different angles and widths of stripes. They also see different parts of the pattern. Each cell only sees an area roughly the size of a one yen coin. So we need to identify which kinds of stimuli are most effective for each cell one by one.
We have developed a reasonable understanding of what cells see in the primary visual cortex. I hope now to advance this research to the higher levels of second and quaternary visual cortex. Some cells express primitive features, so surely there are others that consolidate those expressions together as a group. I want to find out what kinds of forms and elements these higher-order cells respond to. I think that if we can work out a way to measure their responses directly, we will be able to make sense of a lot of other things as well.
Electronic engineering and neuroscience
I studied electronic engineering at university, and only came across neuroscience by chance while I was studying overseas. At first I was just interested to see researchers recording how cells respond. I noticed that when the cells were given regular stimuli, they responded in a regular manner, running contrary to my image of the brain as something mysterious. My interest grew as I realized that this kind research was not so far removed from my own.
Electronic engineering deals with technology, while neuroscience is about life. Some may think that they are very different, but that is not true at all. I think their two worlds are actually rather similar. It is quite natural that someone interested in televisions, cameras, and other image processing technology would also think about how the human brain views the world.
Using a new vocabulary to understand a new field
The Humanware Innovation Program allows, for example, someone who likes physics and computers to do research on life and organisms. What such a student needs to be careful of, however, is not to define themselves simply as a physicist or an engineer and draw the research too deeply into the areas they are most familiar with. I think it is best to put your own prior studies to one side and dive right in to someone else’s world.
As your research progresses, you’ll find that what you learned as an undergraduate student is far from useless. To begin with, though, you should try to express yourself using the vocabulary of the new discipline you are engaging with, and to understand things using that vocabulary. I think that when exploring ideas and expertise from a different discipline, it is important to study the way experts in that discipline talk about it. You have a great opportunity to equip yourself with fresh ideas. I encourage you to make the most of it.