darb.ketyov.com

Caveat lector: This blog is where I try out new ideas. I will often be wrong, but that's the point.

Home | Personal | Entertainment | Professional | Publications | Blog

Search Archive

Loading...

21.1.11

Top 10 neuroscience TED talks

Neuroscience and neuroscientists seem to be a popular staple of TED. As a neuroscientist, this makes TED basically brain-porn for me. There have been so many excellent TED talks on the topic, and I wanted to collect my favorites here (with some commentary). Mind you, this list isn't complete, it's just representative of my neuroscientific tastes.

I've listened to or watched hundreds of TED talks. But I know it's popular right now to give some TED hate. The most common criticism I've seen is that TED is a way for the rich to pat themselves on the back.

But you know what? I've genuinely learned a lot from watching the TED talks. Who cares if it costs folks $6k a pop to attend, when so many talks are posted online for free? In a society where 30 seconds of advertising during the Super Bowl costs $2.6M and American Idol draws 20 million weekly viewers, I'll definitely take some rich folks paying money to gather and talk to each other about interesting topics.

In 2010, I had the honor of giving a talk at TEDxBerkeley about my neuroscience research and my experience growing up watching my grandfather deteriorate from Parkinson's disease. In preparing for that talk, I got a lot of speaking inspiration from some of these talks. Here are my favorites.

Oliver Sacks

There's not a lot I can say about Sacks that hasn't yet been said. His books were a huge inspiration for my career, my research, and my way of thinking about the brain. The profiles he gives of his patients are fascinating insights into how the brain works, and very thoughtful and caring. Seriously, if you haven't yet, go read The Man Who Mistook His Wife for a Hat and An Anthropologist on Mars.

He also just seems like an interesting guy. He did a lot of drugs back in med school, has recently survived cancer (as profiled beautifully by Steve Silberman), and is a regular on RadioLab, my favorite podcast.

In this talk he profiles several patients who have experienced hallucinations caused by damage to the eyes, such as from macular degeneration. This phenomenon is called Charles Bonnet syndrome, and Sacks talks about it in his usual amazing manner.

Jill Bolte Taylor

My first introduction to Jill Bolte Taylor was through her book, My Stroke of Insight. I was given her book by Prof. Marian Diamond, who has become quite well known for her YouTube videos on anatomy. At the time, I was the graduate lab instructor for Marian Diamond's neuroanatomy course. Dr. Diamond knew I working with patients who had stroke, and she gave me Dr. Jill Bolte Taylor's book in the hopes that I gained some empathic insight into what the experience of a stroke was like. She was right. It was amazing.

Now, while I don't agree with all of Jill Bolte Taylor's interpretations of the neuroscience in her talk (e.g., "the right hemisphere 'thinks' in pictures; "the left hemisphere functions like a 'serial processor'"), it's quite amazing never the less.

Henry Markram

Okay, okay, so I've given Henry Markram and the Blue Brain Project some shit on this blog before. Hell, I'm quoted in the New York Times as a bit of a naysayer for this kind of stuff (although my full quote was, "...every neuroscientist will agree that the endeavor is important and worthwhile. It's a necessary tool in the neuroscientific repertoire. The backlash is against the hype.")

But Henry Markram gives a great talk about why brain simulation is important for neuroscience, and he gives one version of how we can go about it. My favorite quote from this talk is toward the beginning when he says, "we can't keep doing animal research forever". While I believe there are also potential future issues with brain simulation, there's a great quote from Bill Crum in his correspondence to Nature:

To my mind, there is a moral inconsistency attached to studies of higher brain function in non-human primates: namely, the stronger the evidence that non-human primates provide excellent experimental models of human cognition, the stronger the moral case against using them for invasive medical experiments. From this perspective, 'replacement' should be embraced as a future goal.

Vilayanur Ramachandran

Ahhhh VS Ramachandran. He's such a great speaker. Sure, he's on record for equating mirror neurons to the discovery of DNA (sigh...), but like Sacks he has some great insight into the damaged brain and, more importantly, what that means about what it means to be human. The Capgras delusion is so fascinating, and I love his explanation of it. Hell, we use it for our fake-science explanation of why zombies don't attack one another!

Anyway he does go on to talk about his classic (and amazing) phantom limb research. Check out his book, Phantoms in the Brain. (Man, I should be getting kick-backs from Amazon for all these friggin' book links.)

Sebastian Seung

I AM MY CONNECTOME! ::chuckles:: Okay, that awkwardness aside, I love his work, and Seung really manages to explain, in clear language, why this work is so important.

Granted, I'm also biased about this one, too...


brainSCANr!


Bradley Voytek brainSCANr


Gero Miesenboeck

Man did Miesenboeck get overshadowed by Karl Deisseroth or what? Nature named optogenetics the Method of the Year in 2010 not because of Miesenboeck, but because Deisseroth had 10 (TEN) Nature* publications in 2010 alone (and he had 4 in Science, too).

Anyway, Miesenboeck's talk explains optogenetics and its applications beautifully. Such a great talk. It also contains one of my favorite TED talk quotes: "So it seems the only trait that survives decapitation is vanity".

And mark my words: the Nobel Prize in Physiology or Medicine in 2020 will be awarded for the optogenetics work of Lima, Miesenboeck, and Deisseroth.

Michael Merzenich

I've got a huge amount of love for Merzenich's research. He's done groundbreaking work showing how important neuroplasticity is for cognition, behavior, and learning. In this talk he manages, in 20 some minutes, to give a very thorough overview of this great work. Again, this is another researcher who greatly influenced my thinking and indirectly lead me down the path to my Neuron paper.

Jeff Hawkins

My views on this talk are similar to how I feel about a lot of the neuroscience TED talks: I don't necessarily agree with all of their arguments or scientific points, but Hawkins offers some cool theories and he's working hard on them in an interesting way. He's got a couple of phrases he uses that I don't like (e.g., "old/alligator brain"), but hey, it's a lay lecture. I'll cut some slack.

I wrote a bit about Hawkins, his book On Intelligence, and Numenta before on Quora. Rather than write it all again from scratch, here's the full thing:

When I first started my PhD, Hawkins was still running the Redwood Neuroscience Institute, which at the time was affiliated with UC Berkeley (since the founding of Numenta, it has now been fully absorbed into Berkeley as the Redwood Center for Theoretical Neuroscience). The RNI was founded on the ideas presented in On Intelligence. The energy of the whole endeavor was amazing, and it was hard not to believe in him.

The main idea in the book is that there are nested hierarchies of cortical modules that give rise to a predictive functionality, and that this is a critical, core functionality of the human brain. There's been some cool research out of Berkeley (e.g., Badre et al., 2009) showing that there is a hierarchical organization within the frontal cortex related to cognition. The idea isn't an old one, but Hawkins organizes it and gives it a good foundation.

It's an excellent start (clearly the brain does make predictions), but it's well-known from psychology that another thing we humans are good at is adding a post hoc narrative explanation to something that we did unconsciously, or that has no obvious explanation. Obviously this kind of phenomenon indicates a "broken" prediction mechanism where we make a "prediction" of something after it has already happened, and then we tend to remember the event as though we accurately predicted it beforehand!

So basically, yeah. Brains is hard. Hawkins is smart and he's onto something, but it's not the whole story. But I'm sure he'll get something out of it in terms of working classification and prediction algorithms, even if those algorithms don't have anything to do with what the brain turns out to be actually doing.

Christopher deCharms

So when I first saw this back in 2008, I was thinking, "ugh, so much hype". Well, here we are only about 3 years later and I've seen more of what Christopher deCharms and his company Omneuron have been up to, and I gotta say I'm a bit more impressed now. It's fMRI-based (and if you know me, you know my feeling about fMRI), so there's that. But the potential for real-time fMRI paired with biofeedback for patient treatment is enormous, so I'm gonna hold my breath and hope they pull this off.

Dan Gilbert

Okay, okay, so it's not strictly neuroscience per se, but damn if Dan Gilbert's talk on how context shapes our behavior and psychology isn't great. Just watch it.



Daphne Bavelier

This is such a great talk providing good, real evidence to counter a lot of nonsense, reactionary claims about the negative effects of video games on our brains. Check out my full write-up about this talk here.