neurodudes

(more…)

NSF:ENG:EFRI:Home Page

NSF’s Emerging Frontiers in Research and Innovation (EFRI) office funded 4 very futuristic neuroengineering grants.

1. Deep learning in mammalian cortex
2. Studying neural networks in vitro with an innovative patch clamp array
3. Determining how the brain controls the hand for robotics
4. In vitro power grid simulation using real neurons

Disclaimer: I was involved with the second proposal on this page.

The Circadian Clock in the Retina Controls Rod-Cone Coupling (Christophe Ribelayga, Yu Cao, and Stuart C. Mangel)

An amazing paper from Neuron demonstrating adaptive (circadian clock-governed) binning in the retina, based on dopamine modulation of gap junction (electrical) synapses between retinal photodetectors. During the day, abundant dopamine release weakens gap junctions coupling rods and cones together so that visual acuity is high. When light is scarce (at night), there is less dopamine and the electrical coupling between rods and cones is increased. This is analogous to on-chip binning in CCD (digital) cameras. Binning increases signal (in light-limited systems, eg. seeing at night) by increasing optical input area and by reducing single element noise (ie. noise at different photoreceptors should be independent) at the cost of resolution. So, the retina activates photoreceptor binning at night to boost low-light signals and deactivates it during the day to increase resolution. The dopamine comes from cells in the interplexiform layer, whose dopamine release is itself governed by melatonin projections.

Also, I never knew that gap junction strengths were directly modifiable. It looks like the D2 receptors are G-protein coupled to PKA, which acts on the gap junctions.

Stanislas Dehaene, Véronique Izard, Elizabeth Spelke, and Pierre Pica. Log or linear? Distinct intuitions of the number scale in Western and Amazonian indigene cultures. Science, 320(5880):1217–1220, May 2008.

The Mundurucu are an indigenous culture whose language does not contain exact words for numbers above 5. Dehaene, Izard, Spelke, and Pica basically gave subjects an empty horizontal line and then gave them a bunch of numbers and told them to place the numbers where they belong on the line (the line was not entirely empty; the left and right hand sides were labeled with a small number and a big number).

Western adults tend to place the numbers linearly, whereas the Mundurucu tend to place them logarithmically. Western children also place them logarithmically, and even Western adults place them logarithmically if the numbers are presented as larger numbers of dots (10-100) or if they are presented as tones.

This suggests that there is an innate logarithmic representation of number which is used to populate the number line, but that with training a linear representation can be created.

From March. Actually, the topic of the article is Dehaene, but it talks about some studies too. Excerpts after the break, interspersed with hyperlinks to citations that I looked up.

http://www.newyorker.com/reporting/2008/03/03/080303fa_fact_holt?currentPage=all

(more…)

Findings: Tapping Into What a Deer Sees, and Doesn’t

Not being a hunter, I’m not sure how much I support this, but I must admit this is at least a very interesting application of psychophysics data. Using deer as subjects in a standard battery of visual psychophysics tests, researchers have engineered a new material/pattern (“Gore Optifade”) that is superior to standard camo for evading detection by deer. Looks like deer are red-green colorblind but have higher acuity in the blue end of spectrum than humans.

Once they had assessed the deer’s visual strengths and weaknesses, Dr. Neitz and Dr. O’Neill worked out colors, textures and shapes with Guy Cramer of HyperStealth Biotechnology, a company that designs military camouflage. Mr. Cramer’s computer algorithms create fractal patterns that exploit a couple of ancient tricks used by animal predators.

The first and most obvious trick is to fade into the background, as a leopard’s spots enable it to do while it’s patiently waiting to ambush a prey. The spots aren’t shaped like leaves or branches, but they form an overall “micropattern” matching the colors and overall texture of the woodland background.

That trick, though, won’t work for a predator on the move, which is why a tiger doesn’t have spots. It has a “macropattern” of stripes that break up the shape of its body as it’s stalking or running.

There is a nice demonstration image with the article showing the same scene viewed with human vs. deer vision.

As the first presidential debate nears, there’s a lot of excitement (and worry) regarding the election. Today, Salon had an interesting piece on voter behavior and irrational attachment to ideologies and candidates. Recounting a recent psychology paper’s punchline:

The article’s conclusion should be posted as a caveat under every political speech of those seeking office. And it should serve as the epitaph for the Bush administration: “People who lack the knowledge or wisdom to perform well are often unaware of this fact. That is, the same incompetence that leads them to make wrong choices also deprives them of the savvy necessary to recognize competence, be it their own or anyone else’s.”

Slate had a story (“Why is every neuropundit such a raging liberal?“) about how neuroscience and neuromarketing are changing political consulting (also here’s a link to a similar story in NYT last week):

According to a study of political psychology published last Thursday in Science, conservatives tend to be the jumpier lot.

The researchers called 46 political partisans into their laboratory at the University of Nebraska, affixed electrodes to their fingertips and eyelids, and measured sweat output and eye blinks in response to a series of startling stimuli. (Subjects were forced to endure images of bloody faces and maggot-infested wounds, as well as sudden blasts of white noise.) The results: Social conservatives—those who supported the death penalty, the Patriot Act, prayer in school, and the like—sweated more, and blinked more intensely, than the liberals.

The Slate and NYT articles in particular suggest something that I have long believed to be true. The Republican “story” is, from a neuroscience perspective, simply better because it tends to view the world in clear-cut terms with no middle ground and, thus, is more effective at rallying emotional processing areas of the brain (eg. limbic system). It is well-known in neuroscience that emotionally salient events that activate these limbic structures are better remembered than less charged memories. The Democratic “story” tends to be more complicated with shades of gray and therefore requires higher-level processing (eg. cortical areas involved in conflict resolution). Clearly, I’m oversimplifying things here a bit (see, I’m designing this post to appeal to your limbic system!) but I think that this hypothesis might have some legs.

Of course, if it’s true, why doesn’t everyone vote Republican if that story is the neurally more rewarding one? Or perhaps the more relevant question: Is it even possible for the Democrats to tap into the similar evolutionarily older limbic structures that seem to dominate the Republican story?

Also, although I prefer Neurodudes to stick with the science over any partisan politics, I must say I found this statistic interesting (from the Slate article):

in 2002, Daniel Klein and Andrew Western tallied the political affiliations of professors at Berkeley and Stanford and found that even in the hard sciences, Democrats outnumbered Republicans by a factor of almost 8 to 1. Among professors of neurology and neuroscience, Klein and Western counted 68 registered Democrats against just six Republicans.

There has been a few articles recently in the NYT about the neural mechanisms used by mosquito repellents. What a wonderful idea: Do some ephys recordings to find which neurons are sensitive to DEET (the current standard for mosquito repellents, which I can attest both doesn’t work very well and eats holes in synthetic clothing) and then build targeted compounds for those receptors/neurons/pathways. I always like this type of simple and practical neuroengineering.

Right now, it appears that there’s a bit of controversy in the field. Earlier this year, in Science, a group from Rockefeller found that DEET masked sensitivity to human odors by interfering with a particular odorant receptor. This impressive result was recently question by entomologists from UC-Davis in a PNAS paper claiming that DEET acts directly on a particular olfactory receptor neuron and does not attenuate the response to the same human-emitted odorant, as found in the earlier paper. Although the results appear to be conflicting, the studies use different techniques and thus it is likely that DEET’s action might be more complex than either paper claims. Still, the idea of identifying a target for chemical intervention by looking at electrophysiological responses to DEET is smart.

In related work, earlier this year a group from Colorado State University, as described in this PNAS overview, “conducted a rigorous search of a library of N-acylpiperidines, using an artificial neural network to identify strong candidates, and then tested them in the laboratory on human volunteers.” They found a candidate molecule that has a ~4X longer repellency effect than DEET. Here’s a photo from the experiments (DEET vs. untreated hand)… ouch!

It has just been brought to my attention that for the last few months (!) neurodudes has had some serious problems with allowing users to comment, ie. commenting was completely closed. I am very sorry for that. The problem should be fixed and now commenting should again be easy for anyone to use. As always, feel free to contact us at contactus^neurodudes^com (replacing ^ as appropriate) if you have any questions or concerns. – Neville

Personal genomics is just starting and this talk gives a preview of what one of the first companies is doing to bring this to market, though the field is starting to heat up with some competition.

Surprising facts from the video:

1. 23andme co-founder Anne Wojcicki is married to Google co-founder Sergei Brin
2. Data portability is already available. According to the video, users can download the raw data from the 580,000 SNP array and do whatever they want with it.
3. The SNP array is 580,000 SNPs! Wow. I assumed that the number of genetic features was on the order of 100-1000, giving a rough haplotype.
4. Sharing/Web 2.0 features: The real power is 23andwe, an effort to data mine and leverage the power of a large database with many people’s genetic information. The founders mention that they want to contribute back to science and healthcare by surveying their customers behaviors and medical issues to uncover further correlations with genes. Like Facebook, the service becomes more valuable and more informative as the network grows.

Description of the talk from Google:

The 23andMe Personal Genome Service offers customers a glimpse at their own DNA sequence, a 750-megabyte string that functions as the operating system for a human being. Common variations in this code can influence the structure and function of the associated wetware in predictable ways. Some of these variations and their effects on traits such as athletic talent, pain sensitivity and avoidance of errors will be discussed in reference to three well-documented examples.