home | about |

as you may have guessed, i’m on another blogging hiatus. they’ve been happening all too frequently, i admit, so i probably won’t start writing again till i have a glut of accumulated posts and time. enjoy the holiday season!

(warning: if you’re not too interested in neuroscience, ignore bohemian scientist for the next week! check here for a good discussion of halloween-eerr-reformation day, and here for answers of high profile academics on the question “is there a purpose?” (courtesy of millinerd).

throughout the next week, watch for updates here on the society for neuroscience meeting in san diego… mainly on the following topics:
sensory encoding (see especially the barrels satellite meeting, thursday-friday)
motor control, satellite conference on friday, assorted posters throughout the week
sensorimotor integration (a few poster sessions and a symposium)
computational systems issues (throughout the week)

and, of course, a good amount of extracurricular work…e.g. all are welcome here:

the society for neuroscience meeting is coming to town! i’ll be covering some it and a satellite (barrels). watch for updates; i’ll post my itinerary soon.

many thanks and kudos to all of the people who helped avert disaster in the wildfires that ravaged greater san diego over the past week. in an attempt to save everything of value, the bohemian server was unplugged and packed to go. fortunately, none of the la jolla areas burned, but bohemian scientist was down for a few days… thanks for understanding!

enjoy these two sites, full of well-known illusions. i’m going to wash out my eyes and try to regain a sense of reality.

a debacle’s a-brewing on copied work in research publications. last year, some software was unveiled that checked papers on the physics archive (www.arXiv.org). the researchers showed just 0.2% cases of plagiarism (assuming their results scaled for the small sample size they analyzed). here’s the review.

last month, a similar analysis on arXiv turned up some 70 offending papers, mostly from turkish theoretical physicists (original story). forty alone were “coauthored” by one graduate student, who apparently was first flagged because he wasn’t able to answer simple mechanics questions at his oral exam, despite his publication record.

online this week, nature posted to more bits to the story. first, an angry turkish physicist weighs in on the accusations: “Plagiarism? No, we’re just borrowing better English.” second, another physicist suggests implementing a different submission method, in which all new papers would be plagiarism-checked and flagged before being accepted.

it sounds like there are two separate issues here. first, there’s the dire case of students who publish (or even pass) without knowing anything. there shouldn’t be much tolerance for this; it makes all of academia look bad. but second, there’s the slightly more innocuous case of poor english speakers. as the correspondence submitter, a turkish physicist, points out, “Even if our introductions are not entirely original, our results are — and these are the most important part of any scientific paper.” additionally, there are very different views of plagiarism (including none at all) in different cultures. for example, the student who was flagged for his poor oral exam couldn’t be expelled because, apparently, his university had no ethics policy regarding plagiarism!

it strikes me as bad practice to write up results while stealing introductions and backgrounds. but it does point out two different styles of science. some very original work synthesizes different sets of experiments in new ways, leading to new directions of research. these obviously require original background material, because they draw from different literatures. meanwhile, there are experiments in single fields that just beg to be done. usually, people in the field know that the question was there and why it was there. all they need is the answer. one can imagine a forum in which original experiments are inserted into a larger published work as the experiments are performed. At its root, scholarpediahas a form of this that might work on the long term. researchers just insert their contribution or experiment into the larger review article, and the knowledge is thus collected (introduced, contextualized, and reviewed) without extra work for the experimental scientist.

scientists often aren’t good writers. to what standard should scientists hold scientists?

as you probably know, this is the week of the 2007 nobel prize announcements. one field’s winners are announced each night; see here for the results so far. the major players in paving the way for gene knockouts, a staple of molecular biology research, were awarded the prize for physiology or medicine.

interestingly, the creators of a major citation index have been trying to predict the winners for several years now. they’re zero for two this time around, and i have a feeling the winner for chemistry, announced in 6 hours, also isn’t on their list. it’s bad karma to suggest his name, so just trust me when i say “i told you so” tomorrow.

UPDATE: here’s a nice post about the chemistry prize (tip to ah): “Very early tomorrow morning, some lucky biologist will receive a call that he’s won the Nobel Prize in chemistry. Just kidding—it will probably be a medical doctor or a physicist.”

the center for ethics in science and technology, based out of ucsd, has declared this week “neuroethics week.” each day, a high-profile speaker and expert panel discuss a subfield relevant to neuroscience and ethics. monday, hank greely, a law prof at stanford, gave a keynote lecture “A Revolution in Neuroscience: Challenges for Society.” below are some notes from the lecture… read at your own risk! (disclaimer: these don’t necessarily reflect the views of bosci).

if these are issues you’re interested in, come to a coffee hour i’m co-hosting this saturday afternoon; feel free to email me for details.

1. ethical issues in neuroscience research
• as many as 40% of participants in research fMRI scans (typically undergrads) have some abnormality. what should the researcher do?
• identification of anonymous research subjects (a problem similar to those in genetics)
• dual use problems. similar to biological warfare today, or nuclear physics of fifty years ago. e.g. monkey stimulation of thalamus can keep people awake. this could be used by our military, or others.
2. the neuroscience of ethics
• the trolley problem: flipping a switch to kill five instead of 20 people, throwing someone in the way to kill just one person instead of 20 people (researchers like josh greene).
• by extension, neuroeconomics and advertising
3. ethical, legal, social implications of neuro applications
• prediction. discrimination for insurance. examples:
• similar to the genetics diagnosis problem (e.g. huntington’s disease); amyloid plaques (indirectly visible by PET or MRI) to compute likelihood of dementia onset (long-term studies are underway).
• schizophrenia
• psychopaths, greater likelihood in crime (1-2% in general population, 30-70% of incarcerated population)
• mind-reading.
• currently, we can read whether someone is looking at a face or a house (fusiform face area).
• we can detect the feeling of pain by fMRI (useful for law and insurance purposes), and could examine the pain associated with lethal injection.
• lie detection. so far, 12 peer-reviewed papers. two companies are doing it: no-lie mri, based in san diego. the relevant research found 80% accuracy. much of this is funded by the intelligence community (darpa, cia, homeland security).
• competence, consciousness, and responsibility
• when a crime is committed, to what extent is he culpable. put differently, is there free will?
• is an elderly person in a capable state to sign a will?
• vegetative states. in one of 13 cases, a patient showed all the same brain activity patterns that a normal person would. what should be done in this case?
• treatment
• try to distinguish the good from the bad. we have a track record of mistakes (the inventor of the frontal lobotomy received a nobel prize long before the technique was discredited.
• a cocaine effects-blocker. should it be mandatory?
• enhancement
• making people “better than healthy”. improved memories.
• brain-computer interfaces (as for quadriplegics, or the cochlear implant, which has been around for 25 years. see michael chorost)
• beta blockers to prevent deep emotions following traumatic stress. it’s already being used in the military and for rape victims.

this neuroethics week is essential to educate the public not on the answers, or even the questions (which depend on the technology development), but on the class of questions we do and will face.

(and an important point during the discussion)
how good must a test be to use it (false positives and and negatives)? should this be administered by FDA?

for those in the area, james watson, of double helix fame, just finished a “public conversation” at the salk, and will be speaking at d.g. wills books in la jolla tonight at 7. according to the link, he may discuss his plans of “making all girls pretty.”

watson recently published Avoid Boring People: Lessons from a Life in Science (available for pre-order on amazon). if you think of attending tonight, remember the title’s deliciously double entendre! (photo credit: max gerber, la jolla village news)

the salk just announced a january symposium called biological complexity: genes, circuits, and behavior. it includes such nobel laureates as richard axel, eric kandel, and susumu tonegawa, along with other heavy hitters like karl deisseroth, rusty gage, tom jessel, edvard moser, and ranulfo romo. place your bets!

see here for details. and thanks to baseball fan and neural star ba for the tip.

first, rather than explain yet another virtual absence, i’ll point you to some things that have unapologetically stolen my time. with any luck, you’ll get lost in at least one of them, and forget that you haven’t read anything here for a while.
{google reader; tiddlywiki; enc of philosophy, eg the problem of induction; the ugly insides of data acquisition; etc}

ny times, glynis sweeny

at any rate, nytimes had a much-needed op-ed by mike males, a guy who helped to start youthfacts.org. the article and the site try dispelling myths about youths. males makes several interesting points as he demonstrates a “ballooning crisis” among americans aged 35-54 in traditionally adolescent issues like drug and alcohol abuse. beware of his numbers comparisons, though, such as this one: of all 35-54 years old americans, “21 million [are] binge drinkers (those downing five or more drinks on one occasion in the previous month), double the number among teenagers and college students combined…” here, the adult group range is twenty years, whereas “teenagers and college students” are an age-group spanning only ten years. that is, the population size is probably about double, making roughly the same percentage of binge drinkers in the two age groups.

what i especially appreciated about males’ piece, though, was his willingness to challenge claims about youths that are traditionally made on the basis of irrelevant or misrepresented research studies. i’m not the first to complain about this: the media’s science fetishism, mixed with its sound-byte sensationalism, render impossible any attempt to fully gather data and make fair interpretations. of course, the answer isn’t to downplay the significance or applicability of research… perhaps just to rein in the lay coverage of its conclusion.

for comparison, see also this interesting scientific american human behavioral and fMRI result regarding political disposition and internal conflict. here’s a quotation from the article, which was a review of a recent nature neuroscience paper: ” ‘They are more sensitive to the need for change and more sensitive to the need to change their behavior,’ Amodio says about the politically left-leaning subjects.”

credit: blog carnival

welcome to the twenty-eighth edition of encephalon, a circus of recent highlights from the neuroscience blogentsia. this time around, we had many reviews of some interesting original research, along with posts on everything from aesthetics to eulogies. enjoy!

neo-neuro fields

one of the beauties of neuroscience is its universality: at some level, everything involves the brain. too often, though, people affix “neuro-” to the front of their favorite subject, then claim victory over a paradigm-shifting new discipline. two blogs dealt with this issue recently: neuroaesthetics led to some deep insights, whereas “neuro-leadership” just fell flat. both posts were entertaining and insightful.

over at the third culture, jon follows up his two part series on neuroaesthetics with a post on art, context, and the brain. he asks with subtlety, “if we are to believe that there is some way to understand reactions to art by understanding the brain (or to understand the brain by understanding art), how are we to incorporate context-specific reactions?” jon takes on the question with a review of some neuroeconomics and an apropos reference to ucsd neuro-rocker ramachandran.

meanwhile, the trusted advisor asks “is neuroleadership more than reinventing wheels?” he decides that neuroscience hasn’t contributed novel insights to business management, despite the claims of people like the prolific jeffrey schwartz. i couldn’t agree more: projects such as “neuroleadership theory” only detract from more legitimate descriptive science aimed at understanding the brain more than making a buck.

credit: david linden

and now for some reviews…
biology and neuroscience

psychology and neuroscience

credit: mindhacks

reviews of books and lives

and in other news…

at sharpbrains, andreas describes how dancing is mental exercise. he created a video on the neural substrates of dance, and goes on to cite a study in the new england journal of medicine showing that dancing, more than any other physical activity studied, is correlated with less presentation of dementia later in life. sudip at brain blogger provided a brief discussion of a recent science article on the role of nmda receptors in dentate cells of the hippocampus. the exciting work was done in the lab of susumu tonegawa, a ucsd grad and nobel laureate. brain in a vat gave a shout-out to america’s nerdiest videos, a.k.a. the journal of visualized experiments. to be sure, it looks less like a journal and more like a youtube with citations. but it seems like a useful, if small, repository for teaching biological methods.

phew! that’s all folks. the next stop is at memoirs of a postgrad in mid-august. as always, submit here. happy posting!

alas, my month-long blogging break turned into an all-out two-and-a-half month sabbatical. but instead of reflecting on all of the advances in systems neuroscience and the interactions among science, faith, and culture since the ides of may, i’ll point to a nice post from the free geek. back in june, he sent me his do-it-yourself guide to becoming a cyborg (maybe i should take a hint?). it’s an entertaining write-up on some of the human-machine interface options, broadly conceived, that recently have come to market or are in the line-up. watch out especially for RFID implants to gain nightclub admission and nerve implants that record and input sensory signals (that’s right, you can now experience someone else’s peripheral sensations!).

in other news, the sloan-swartz center for theoretical neurobiology at ucsd is currently hosting its annual meeting, featuring a who’s who in neuroscience theory. drop by if you’re in the area.

be sure to stop back here tomorrow, when the twenty-eighth encephalon blog carnival will perform at the bohemian scientist’s shop. for the last encephalon edition, just so you know what neurific excellence to expect, see here.

{picture credits: free geek (1) and sloan-swartz (2)}

thanks to some pressing responsibilities previously shirked, i’ll be checking out of the bohemian shop for about a month. at the suggestion of a reader and close friend, here’s a “best-of” so you can wistfully pine for bygone days. i’ll be back on june 16!

remember the day when the great nyc,
curled into a fetus with no-trans decree?
or when the true godhead was thought to be three:
the mother, the child, and the womb died for me?
or when we found e-fields that drape the body?
why praying is good even for bourgeousie?
how brain drain can benefit any country?
the way to fix eggs broken by G.O.P.?
and tiffs among brain-geeks who cannot agree!?

if coding to europe seems too far-fetched, matlab today released their twice annual competition. you have a week to optimally solve the peg-jumping problem with the leanest code. the winners get bragging rights, and maybe a coffee mug. (thx to bc for the tip)

gerstner’s group in lausanne, switzerland has announced a competition to predict the electrical behavior of individual neurons in two respects:

1) predict the timing of every spike that a neuron emits with a precision of 2ms.

2) predict the subthreshold membrane potential with a precision of 2mV for arbitrary input.

details on the competition, including the dataset (released 16 March 2007), are here.

note that the first prize receives:

- 4 nights of hotel in Lausanne at the Lake Geneva, June 23-27.
- Free participation in the Quantitative Neuron Modeling workshop June 25/26
- 35-minute-slot for talk as an Invited Speaker in the workshop.

get coding.

on friday, sebastian seung stopped by ucsd and the salk. he gave two talks, one on machine learning and a second on connectomics (see this review). this post won’t be critical. these are notes from his second talk. facts that should be noted before starting are that 1) the experimental work was done by winfried denk and kevin briggman, a ucsd comp neuro grad, and 2) the theoretical work was done by machine learning theorists long ago. also, this is just in-seminar note-taking; there’s been no attempt to make it appreciable to people unfamiliar with the topic (i’ll give you a post soon). enjoy!

• two problems of connectomics (the complete connection matrix of all cells) are
• determining what constitute synapses
• determining to which cells each axon is synapsed.
• techniques: use diamond knife, image slice with TEM, or (denk) image block with SEM (serial block face-sem… “the denktome”) (denk and horstmann, PLoS Biology, 2004; denk and briggman, Current Opinion in Neurobiology, 2006)
• limitations: staining a block is hard, electrons going too deep in the block (if too high energy) which will blur the z-resolution. so the x-y resolution is less, ~25-30 nm) (consider)
• nematode is example of successful reconstruction, but done by hand.
• near future: teravoxel datasets. 1 cubic mm, entire brains of small animals, small brain areas of large animals (the lore: c. elegans shows no individual diversity.)

• accuracy- (following neurites: image segmentation (contour completion is easier for humans than for algorithms. which is a bad omen for 3d images, since even humans stink at 3d contour completion); recognizing synapses (object recognition).
• note that medical imagers have somewhat solved this. but the problems they solve are easier. note: 6 sigma imaging (reference to engineering: wikipedia article)
• speed: teravoxel datasets

• NICE: briggman came up with a stain that only stains the membrane = contours (HRP), making comp sci problem easier. machine learning approach: train neural network on sample data (contour completion done by hand by undergrads) and generalize to whole set. in computer vision, david marr thought everything could be bottom-up. that failed. high-level knowledge is required. same for speech processing. harvey brings up the importance of light microscopy.
• the neural network details: convolutional network (related to a late 80s concept, the neocognitron, designed by kunihiko fukushima, a bell labs guy).
• network of convolution operations in 3D. progressively different convolutions (filters in feature maps). linear filter, folllowed by sigmoidal “squashing function”. to arrive at the filters, all parameters are adjustable by training (30,000 parameters, using backprop “gradient learning”, as from rumelhart and hinton. (n.b. from dm: see also werbos).
• each layer (from fukushima): reduced the resolution in each layer (like areas of cortex). this yields spatially invariant representations. here, their algorithm has a biological analog in multiple time steps in V1. “unfolding in time”.
• there are unsupervised approaches that one could do. they did this because it was guaranteed to work.
• shows a fly-through of reconstructed neurons to the bladerunner soundtrack. “though we can’t do science yet, we can do movies”. only 20 um

• structure-function relationships in neural networks
• neural development - how the brain wires and rewires itself
• “connectopathies” - subtle neuropathologies underlying psychiatric disorders.
• example. avian brain area HVC (test synaptic chain theory of motor sequences)

• as compared with lichtman’s projectome.

• 30 um slices of raw images. 50,000 petavoxels (1mil for 1 petabytes); connectome is 5 petabytes, which would fit in google’s storage (around 15 petabytes); 10^11 neurons, 37 bit address (5 bytes), 10^4 connetions/ neuron. in a few decades of moore’s law this’ll work. but this will require nanelectronics.
• terry’s less optimistic that even in decades they’ll be able to do this. only harvey karten can do 3d intuition… the rest of us are not evolved for it.

References.

1. K. Fukushima. Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position. Biological Cybernetics, 36(4): 93-202, 1980.
2. Denk W, Horstmann H. Serial block-face scanning electron microscopy to reconstruct three-dimensional tissue nanostructure. PLoS Biol. 2004 Nov; 2(11): e329.
3. Briggman KL, Denk W. Towards neural circuit reconstruction with volume electron microscopy techniques. 2006 Oct; 16(5): 562-70.
these icons link to social bookmarking sites where readers can share and discover new web pages.
• 
• 
• 

the village voice, an internet daily focusing on nyc culture and politics, serves up “a graphical dissertation on the number one song in america,” rap artist mims’ inimitable this is why i’m hot. when i first heard this song, i could actually hear my own frontal cortical neurons committing suicide. it confirms the death knoll sounded in idiocracy last year and in george orwell last century. (thanks to vt and bm for the tip. the figure is modified from vv’s figure 3).

from the article:

The other remarkable, oft-quoted line in “This Is Why I’m Hot” is “I could sell a mil’ sayin’ nothin’ on a track.” Critics gibe that “This Is Why I’m Hot” proves precisely that; others muse on what Mims would sell if he deigned to actually say something on a track. Would he sell less than a mil’? Exactly a mil’, as when he said nothing? Or a great deal more than a mil’? The song does not elaborate.

the flapperome

it’s not news that scientists aren’t all linguists. but that ain’t no excuse for the nominal slop that pervades biological appelation. perhaps the most notorious of cases, and the subject of today’s exposé, begins in the 1920s: flappers were a dime a dozen, histologists were still arguing over the neuron doctrine, and a forty-something german botanist was coining the term genome. with it, he unwittingly hacked open a pandora’s box of lingual lament, unleashing the phoneme-turned-meme -ome to a world poised to beat it to death. the inevitable ome-philia is now as annoyingly crazy as its shakespearean counterpart; one can only wish it were as self-destructive.

an ome-ome

these days, new *-omes appear daily. perhaps, as colleague dh pointed out in conversation, it’s due to the success of the human genome project, especially as compared to its drunk-in-a-gutter step-brother, the human brain project. here are a few of its more colorful manifestations: the human cytome project, the human epigenome project, the human GNU-ome project, the human Jen-ome project, the american meme-ome project.

the range of possibilities is much greater when it’s not a “human project”: the proteome (coined by mark wilkins, 1995), the metabolome (coined by oliver et al, 1998), the interactome, the transcriptome, the davematthewsome… the list goes on. what spawned my own expl-ome-ation was the new fix that the neuroscience community has with omomics… presenting [drum roll please]… the connectome (PLOS biology article and the obligatory dotcom attempt). it begs the question: what’s in an -ome? surely, people recognize how passé (and strange) it is to affix their favorite subject’s moniker to the -ome suffix: most of these things have nothing to do with chromosomes. but maybe that’s not the real origin of -ome…

et-ome-ology

in the 2 april 2001 edition of the scientist, lederberg and mccray offered a brief but entertaining account of the etymology: ‘ome sweet ‘ome-ics. the authors are iconoclastic enough to suggest that OED gets it wrong. OED suggests that botanist hans winkler, who was the first to use the term in 1920, coupled “gen” from gene and “ome” from chromosome. meanwhile, lederberg and mccray point out three other intriguing, if not all etymologically plausible, facts. 1) there are a number of botanical terms that predated winkler and his cronies in the 20’s, and may have provided more proximate motivation than the term “chromosome”; 2) in Greek, the suffix denotes “having the nature of” ([2], by way of [1]); 3) in Sanskrit, it means “fullness [and] completeness” [1]. beautiful, but winkler was a german at a time when indians were brits… and nobody likes brits.

in the end, skill and passion go hand in hand with the best of science, no matter how it’s dubbed. in that vein, perhaps the old aphorism says it best: ‘ome is where the ‘art is.

UPDATE: see notes from a recent seminar on connectomics (in case you’re still interested).

also see:
[1] Lederberg and McCray. ‘Ome Sweet ‘Omics–A Genealogical Treasury of Words.
[2] Roland Brown. Composition of Scientific Words.
[3] Walter Flood. Scientific Words.
[4] Sporns et al. The Human Connectome: A Structural Description of the Human Brain.

gateway electronics is closing its san diego doors after many years of faithful, if disorganized, service to the techies of SoCal.

our lab scavenged the place a few days back, mostly for personal projects, and walked out with easily more than $1000-worth of stuff, all for around fifty bucks. some of the more exciting purchases were multiple lasers, including this helium-neon one, which had a used pricetag of $100. i got it for one dollar.

i also bought near 100 7- and 11-segment LEDs, you know, just in case… you ever need to build an alarm clock from a couple of watch batteries, a paperclip, a rubber band, and a few LEDs.

the closing marks a well-recognized, but nevertheless sad, shift in our culture: people are increasingly using sophisticated equipment while avoiding the knowledge of how the heck any of it works. as a famous neuroanatomist, personal hero, and close lab friend emailed me regarding the event:

I guess I am of a different generation. No, I KNOW I am of a different generation. When I was a kid in NYC, the area around Hudson Street had dozens of stores that sold electronic components, army surplus transmitters and receivers, 807 vacuum tubes, 6au6 pentodes (?), and transistors hadn’t yet been invented. (Shit! Am I really that old?! Had they yet invented resistors?).

Sad day, and the end of electronics hobbying in San Diego. Hmm!! Maybe that explains why most of today’s kids don’t have the slightest idea how their various electronics toys work.

fortunately for tinkerers everywhere, the head salesman is planning to reopen the store under new guise nearby. plus, their website will still ship out of st. louis. phew!