More videos from the Redwood Center for Theoretical Neuroscience

Charles Anderson: A Comparison of Neurobiological and Digital Computation is a talk given at the Redwood Center for Theoretical Neuroscience, UC Berkeley on 04/10/06, now available on video. From the abstract:

...a computational framework [1] that shows how large networks of spiking neurons can store and transform analog signals for sensory processing, motor control, and statistical inference. The resulting computational systems differ from traditional artificial neural networks that are focused on the highly nonlinear properties of individual neurons, and are more in line with modern Bayesian systems. The brain is more like an analog computer than a digital one; more like a Bayesian inference machine than a symbolic one.

Out of my depth, but intriguing, as is Jack Cowan's talk: Spontaneous pattern formation in large scale brain activity: what visual migraines and hallucinations tell us about the brain.

You may recall the new Redwood Center, opened in 2005, placed a symposium's worth of videos online. It's fabulous they're webcasting regularly. Search the Internet Archive for all their releases. (As of May 12, 23 titles, freely available and distributable through Creative Commons).




Philippe Van Nedervelde's E-spaces 3D Art Studio created Nanobots replacing neurons (nerve cells). This a still image of HD animation, depicting robots in the brain, from the transhumanist documentary Beyond Man (2005).

Although imaginary, current developments like MIT's nano neuro knitting experiment in which peptide nanofiber scaffolding was used to repair brain tissue in a hamster (article in PNAS) make it prescient.

Van Nedervelde's art is part of NANO, a virtual exhibit curated by transhumanist visionary Natasha Vita-More to showcase nanotech art by artists from a number of disciplines, including the multitalented Anders Sandberg. He's a computational neuroscientist and research associate at Oxford's Future of Humanity Institute.

Listen to a recent interview on the subject with Natasha Vita-More by RU Sirius from The MondoGlobo Network, as well as a podcast called Global Design and Nanotechnology from the (podcast archive alert!) Futures Podcast on Culture and Technology series.

[Also-must-see tangent: ZERO@WAVEFUNCTION: nano dreams and nightmares. Superb interactive nanotech art.]



Prizewinning neuroart

Science illustrator Graham Johnson won first prize in the 2005 Visualization Challenge from Science magazine with his beautiful image The Synapse Revealed. Based on brain slice microscopy, he used a combination of pencil drawing and computer manipulation to produce the final result.

The resulting image is a careful balance between precision and beauty. Because the original data were so complex, Johnson cut the number of neuron interactions depicted to only 30% of the original data--"otherwise, it's just a mass of spaghetti in front of you," he says.

"It gives us the information we need, but at the same time brings an aesthetic, a refinement," says panel of judges member Felice Frankel. "That's really important: to get the viewer to want to look--and then to ask questions."

Read more



Interview with Steven Hackworth

Steven Hackworth is a Ph.D candidate at the University of Pittsburgh who worked on the DBS-RF (a wireless deep brain stimulator) as well as the Radio Frequency-powered Neural Stimulator (RFNS) vagus nerve stimulator system. He's now doing research on energy sources and medical implants. After writing about RFID in neurotech the other day (a prequisite to this post, many links there) I did a brief email interview to follow up.

Are the DBS and VNS RFID systems using the same principle or technique or were there unique challenges for the two types of devices? I'm wondering how portable the technology is to other neurostimulation and medical devices, essentially.

The DBS and VNS systems do use the same RF technology. Magnetic inductive coupling is used for powering and communication. There are two major differences between the devices though: 1) The DBS device uses voltage pulses while the VNS device controls current injection. It was just a matter of modifying the circuitry on the output side to adapt the DBS device for VNS. 2) The implementation of each device must be different. i.e. the DBS device is under the scalp, requiring an external powering unit on the head, while the VNS device is located in the neck, requiring an external powering unit in a person's collar or similar. Though I did develop an example of a powering system (with the primary coil in a hat), this issue wasn't investigated thoroughly, and would be left to a company to come up with a suitable implementation (is a person supposed to wear a bag on his head whenever he takes a shower? ... etc.). Back to the essence of the question, the same fundamental RF technology should be transferable to other neural devices.

How do the systems ward against RF interference [thanks BGP for that question], and potential hackers? Is there a designated frequency for medical devices, different from retail RFID and such?

The inductive coupling mechanism used has a fairly short range. Depending on the system, it might be up to a foot, but practically, you don't want to blast people with RF energy. The current systems for DBS do in fact use RF communication for programming (Medtronic's system for Parkinson's, at least), but the user on/off switch is controlled by a magnet, making it susceptible to magnetic interference. Using an established protocol to turn the device on or off would guard against unwanted toggling. I think Medtronic's newer systems may be implementing this, as I've seen newer designs for the patients' controllers, but I can't say for sure. Regarding hackers, I doubt that is an issue. I haven't heard of any malicious cases
with the current easily controlled systems, so I doubt if anyone would go to the length to figure out the communication protocol and turn people's stimulators on or off. The reason for the protocol is to keep the probability of unwanted interference negligible (just like wireless internet and cell phone protocols).

The frequencies used fall into the FCC-designated ISM bands. Industrial, Scientific, and Medical. Anyone can use these frequencies, as long as they abide by certain rules and regulations. There are fairly standard frequencies used for RFID systems, and medical devices could use those or others, as long as they abide by FCC regulations.

In your current research with energy to implants, what are the main challenges (in a nutshell), and promising developments? (i.e. New types of batteries, other energy sources? What about using the body as an energy source, is that feasible?)

The main challenge involved with getting energy into the body is getting through the body itself. It is highly conductive, and conductive layers tend to block any sort of electro-magnetic signals.

However, we are doing research on using those conductive properties to actually transfer signals instead of acting as a barrier. We are also looking into what we can do to modify tissue properties to increase energy transfer efficiency. As mentioned earlier, practical implementations of the technology will be an issue. There is some research into harvesting energy from the body. Though still in its infancy, it looks ever more promising as technology improves and requires less power.

Thank you Steven!



The recent Science article Neurobiological Substrates of Dread, Berns et al. (2006) has drawn a lot of interest. In the mainstream media many similar dry and simplified accounts are propogating, but the blogosphere has some great analysis. Three unique perspectives:

In the Psych Central blog, John Grohol talks about distraction easing dread and notes: "...while the researchers were surprised to find that the anticipation of pain was in the rear of the brain rather than the front, I’m not sure most researchers would share that surprise. Avoidance of pain has been shown for decades to be a very core behavior in mice and humans. Nobody likes pain."

Also, "[The study] really has no impact on people’s current or future treatments when undergoing painful procedures or such in medicine."

Neuromarketing blog does draw a practical conclusion: "Between mirror neurons 'simulating' an observed (or even heard) action by another, and pain centers being triggered by thinking about future pain, it’s clear that marketers may have the opportunity to create discomfort among their targets." He gives icky examples, and warns, "I’d advise advertisers to be aware of the reality of pain anticipation and to use painful imagery with caution."

Meanwhile, The Neurocritic nimbly guts the study: "...the authors commit the logical fallacy known as 'reverse inference' by inferring the participants' emotional state from the observed pattern of brain activity. They discount the role of the amygdala in 'dread' because both moderate and extreme dreaders showed elevated hemodynamic responses there during the unpleasant interval of waiting for the shocks."

Taken together, the anatomical locations of dread responses suggest that the subjective experience of dread that ultimately drives an individual's behavior comes from the attention devoted to the expected physical response (SI, SII, the caudal ACC, and the posterior insula) and not simply a fear or anxiety response.

"So anticipation of pain is 'attention,' not fear and anxiety. It's a little early to make that conclusion."

More good points in the blog comments, too. How about yours?



RFID in neurotech

“You have to void yourself of prejudices, pre-conceived ideas or whether you’ve seen this type of person before when it’s time to get to work on something,” said Marlin Mickle. “Because if you don’t, your mind starts to focus on that and you don’t get to what matters most. Nothing is useless to the man of sense – everything is taken into account.”

Mickle is a professor and director of the University of Pittsburgh's Radio Frequency Identification Center for Excellence and of the John A. Swanson Institute for Technical Excellence. Many innovative solutions have been developed in his labs - including a wireless deep brain stimulator and vagus nerve stimulator. The latter was developed with the help of a group of high school teachers.

The Radio Frequency-powered Neural Stimulator (RFNS) is described:

The RFNS is made up of a receiving device implanted under the skin of the neck and a powering device placed near the skin at the same site, under a collar. Because this requires only one surgical incision, rather than the two required by VNS, the risk of infection is reduced. Other advantages of RFNS over the existing VNS system include no invasive tunneling from the shoulder to the neck region and an external battery, which reduces the need for subsequent surgeries and further lowers the risk of infection.

Mickle told Applied Neurology, "People are lining up for it." The wireless DBS has already been licensed.

The only FDA-approved vagus nerve stimulator, from Cyberonics, has wire leads tunnelling through the neck. Other neurotech devices also use leads, including various DBS systems (for depression, Parkinson's, epilepsy primarily), and one that stimulates neural plasticity following stroke (also being investigated for tinnitus). Northstar Neuroscience manufactures the latter; they're presently recruiting an engineer to work on leads. Why?

One good idea can suddenly make all the others seem so clunky.

Tags: neurotech RFID future


Podcasts and a PSA

Dana Alliance for the Brain produces a neuroscience podcast series called Gray Matters as well as panel discussions and more. These recordings include thematic collections (neuroethics) and one-on-one interviews (Eric Kandel). Access the complete collection on their archived webcasts page.

Among the offerings:

Depression and Bi-polar Disorders

Prominent faculty from The Johns Hopkins Mood Disorders Center discuss the latest research and clinical findings on depression and bipolar disorders. Featured speakers included co-directors of the Center and Dana Alliance members Kay Redfield Jamison and J. Raymond DePaulo, Jr.

I think Dana does great work and I'm thrilled about this archive and all but *ahem* they hit on a pet peeve.

Bipolar is not two words. You don't ride a bi-cycle to see your BiSexual lover during a Bi En Ni Al. It's bipolar. Just one word. No hyphen. Lowercase. Simple.

This has been a public spelling announcement.



Neurofutures that aren't mine

Neurofuturity: A Theory of Change coins a new meaning for neurofuture (the neuroword, not this blog). It's an article from Clinical Child Psychology and Psychiatry. "Capturing the possible worlds to come is described here as neurofuturity. It is much broader than expectations and includes our feelings towards the future as well as our beliefs."

Another online "neurofuture" that isn't me is this guy, who is spamming the blogosphere selling expensive cognitive brain fitness products. It is most definitely not me, when you're Googling around and finding thousands of hits pushing brain fitness. Rather annoying.

Brain fitness (a.k.a. neurofitness, a neuroword and buzzword that isn't catching on) was debunked in Mental Exercise and Mental Aging: Evaluating the Validity of the "Use It or Lose It" Hypothesis, a review published in Perspectives on Psychological Science that reveals no evidence to support the claim. Not that it's necessarily wrong, but it hasn't been tested using the right approaches (it's all been correlative rather than causative).

Dr. Timothy Salthouse does also conclude, "Despite the current lack of empirical evidence for the idea that the rate of mental aging is moderated by amount of mental activity, there may be personal benefits to assuming that the mental-exercise hypothesis is true."

Maybe that's enough to warrant selling a video game, but it doesn't justify spamming.



Global village hut broadcasting

Via Omni Brain: Neuro Aesthetics -Watch-: videos from a Columbia conference on neuroaesthetics, Art and the New Biology of Mind, are now online. They features talks by VS Ramachandran, Eric Kandel and other huge names.

Bora of Science and Politics also blogged it, objecting that the artists' comments after the talks were apparently the best part but are not on the videos.

Some online lecture videos do include the Q&A; it's a better format since it may answer viewer questions, and as Columbia has now discovered, it'll preclude outcry over perceived censorship. I doubt that was considered. It could be that the webmaster was trying to save bandwidth with shorter clips in smaller files, or maybe thought remarks were irrelevant. Many people would like to know what Laurie Anderson's comments were, though, even if negative. In presenting it online, the aesthetics and marketing of neuroaesthetics are, ironically enough, mostly ignored.

It'd be great if we had skilled journalists covering conferences to get that kind of comprehensiveness: packaged video documentaries covering of all aspects of an event. Not just the lectures; a lot of interesting stuff at conferences comes from informal talk over coffee breaks, and there could be "confessional cams," etc. But these are early days of crappy badly lit RealVideo online, somebody setting up a camera almost as an afterthought, and it seems we're lucky to even be getting that. Most conferences are still offline.

Institutions creating webcasts, and podcasters, could learn a lesson here. New media i.e. the intarwub, is a synthesis of old media used in new ways - learn about the old to more effectively use the new. (Columbia is renowned for its journalism program, and yet they present webcasts like this.)

Oh, and didn't anyone attending that conference have a videocam on their cell phone and a blog with RSS, or a YouTube account? Could be that those post-lecture comments, and reader comments on those, are already online.

That'd be too easy, wouldn't it?


eXTReMe Tracker