요즘 가장 재미있게 듣고 있는 강의. 이미 중반부를 넘어섰지만 지금 등록해도 강의자료에 access할수 있기 때문에..신경과학도들의 must take course로 강추해봅니다. Coursera에서 이뤄지고 있는 브라운대학의 Exploring Neural Data란 강의인데, 일단 강의구성이 외형적으로 매우 고퀄입니다. 인테리어 우아한 커피샵에서 공부하는 느낌이랄까. ㅎ 강의하는 교수님과 instructor 박사님도 강의를 무척잘하구요. 기존의 neuroscience 강의들과 굉장히 다른 focus를 갖고 있는데, 제목 그대로 data가 중심적인 키워드입니다. 콘텐츠의 비중은 비디오강의: 데이터분석 프로그래밍 assignment가 3:7 정도라고 생각이 됩니다. 그만큼 실제 다양한 neuroscience 데이터를 분석하면서 공부할수 있도록 신경을 많이 써놨습니다.(프로그래밍 언어는 Python) 비디오강의는 해당 파트의 그리 복잡하지 않은 수준의 신경과학적 지식 개요, 그리고 신경과학과 데이터 수집, 분석에 대한 이슈들, 데이터테크놀로지의 발전 등에 대한 discussion guest들과의 discussion, 그리고 Python, numpy 강의(해당 주의 과제를 하기 위해 필요한 내용들 위주로)로 구성됩니다. 데이터와 신경과학에 대한 다양한 주제의 디스커션들이 참 재미있는데 지난주 discussion 의 스크립트 일부를 옮겨와 봅니다. ================================================ Q>>if you could look into the future a little bit and, maybe some of the grad students in, in your lab. When they become professors, what are the questions you think theyll be studying? And what technology do you think theyll have to make their lives easier in their research work? A>> I think, thats a good question. I, I suspect, a number of things are going to change some of them driven by technology. Some of them driven by a, a realization that we sort of need to study the, the whole system. ... itll probably be less and less common to find people focusing on just a single area of the brain. which, many of us have done partly because its sort of an entry into the problem. But we all know that theres not, its not as if theres one, you know, magic node down there.That if we understood it, wed understand everything else. The whole system works together, and so, what you have to be able to do, is to look at what everything is doing, as a group. And in order to do that, and what makes that so hard, is you need to be able to do that for. At the same time, if you really want to look at how areas communicate. And so this requires technologies that allow many cells, and many small sub areas to be studied. But you have to be able to do that across multiple areas. And, some of the constraints I think were going to see sort of removed would be some of the limitations. For example, even on movement that, that we would, we should and we will be able to do this in a freely moving animal, for example. By virtue of taking advantage of some of the wireless communication. Having a lot of that electronics be so miniaturized and so, low power that it can just be placed right there on the skull in a way. Sort of like people walking around with pacemakers. This is not uncommon these days. So to shrink that further, and to have all of that, those sensors be available. That means that youd then be able to, think about a task thats not just sort of one small slice. Lets say, how base might affect a neuron in inferotemporal cortex, but how a whole behavior unfolds. How does the information come in and they convert, and then get converted into, one of many actions, and then hows that action actually planned? And if you were recording from multiple areas, youd be able to see this, at one shot. I mean that would be. I, ima, you know I, I often joke that its a, its a little bit strange to, to study a system in an incredibly reduced case. Its like, you know having a Ferrari in your, in your, in your backyard and only telling your friends about how great it drives in reverse. This doesnt make any sense. I mean you know, this car is supposed to do one thing and, and our brains are designed to do something,very complicated and, and, and solve really hard problems. We know it can do this, but we also,if we want to understand how that happens. We actually need to study it in those conditions, and so I think thats the direction well be going in. I think also the data I should say, will change. I, I dont know how its going to happen in terms of streaming. This will be be a lot of data. And youll need lots of hands on deck to make some sense out of it likely. And Im sure that on the analysis side, things will change dramatically as as smarter and smarter, sort of mathematicians are sort of joining the fray and trying to understanding the language of the brain. That may also mean that the data that are kind of flowing out of a laboratory like Ive described, might make its way into you know a, pool a crowdsource-like situation. Where the data are available essentially to all people who are interested and capable of making some sense out of it. And and you know, who knows what itll look like in terms of data sharing. And and, and availability in, in ten or 20 years, but Im sure itll look different than it does now so. Well, we all look forward to finding out, I guess.
Posted on: Sun, 09 Nov 2014 13:34:44 +0000
Trending Topics
Recently Viewed Topics
© 2015