May 11, 2006
This is Why We Need a Subscription to Neuron
Way cool article posted on neurodudes.
If ever there was a paper that would bring tears to one’s eyes, this is it: a previously hidden mental process has now become subject to experimental study. The mental process is the covert movement of attention, the selective focussing of attention to subregions of the visual field, but without eye movement. The movements of covert attention were hypothesized based on psychophysics, but the authors can now follow it using a vector field derived from a population of neurons in the parietal cortex. The monkey has been trained to use covert attentional shifts to solve a maze task. The major finding is that the vector derived from the population of parietal cells follows in time the path through the maze, as the monkey solves the maze.
Posted by todd at 4:46 PM | Comments (1)
February 18, 2006
How to Make Hard Decisions
Post by John Hawks with this for a punchline:
"At some point in our evolution, we started to make decisions consciously, and we're not very good at it. We should learn to let our unconscious handle the complicated things," Dijksterhuis says.Strong statement! How can you not read the rest? Now, where is my subscription to Science?
Posted by todd at 12:10 AM | Comments (0)
February 16, 2006
Weather on Mars, People in Brazil
- Certain areas of Mars have the same weather every year. (Keats' Telescope)
There are specific times of year and locations on Mars which have experienced the same dust storm patterns every Mars year since we began observing with the first MGS MOC approach image in July 1997.
C'mon, that' s just neat. - At Mixing Memory, a post about cultures with few words for numbers. It summarizes a couple of articles from Science about societies in the Brazilian rainforest where they only have words for "one, two and, many" or "five words for different general numeralities." The Saphir-Wharf hypothesis shows up! That's always fun to see. Also, this, which I thought was awesome. Sven would love it:
The picture I'm trying to paint with these two sets of studies is just how messy research on language and thought really is. [...] Figuring out whether language, as opposed to other cultural and/or environmental variables are responsible for differences in cognition is damn near impossible to do with any certainty. But the research is fun anyway.
Posted by todd at 11:05 PM | Comments (2)
February 13, 2006
What does it take to empathize with someone you hate?
Interesting article at Cognitive Daily:
A team led by Kerry Kawakami explored whether priming could affect attitudes which opposed the inclinations of the people being primed. They asked college students to describe a picture of an elderly woman for five minutes -- her hobbies, interests, and general character. Then they asked their opinions on issues which young and old people typically disagree: health care, TV nudity, etc. A control group was primed with a picture of a young woman, and another group wasn't primed at all. In every case, the college students who had been primed with the concept "elderly" expressed more conservative, "older" opinions. Suspecting that the participants might have caught on to their hidden agenda, the experimenters interviewed them to learn if they had adjusted their responses because of the priming; none indicated any awareness of the purpose of the study.
Read the rest to find out how it relates to terrorism and made-for-tv movies.
Posted by todd at 6:12 PM | Comments (0)
February 11, 2006
Neuroscience Reading List
A reading course in computational neuroscience at Berkeley has a webpage up with a really handy reading list of neuroscience and AI papers, for the most part complete with links. Of course, a lot of the links will be useless from non-campus networks, but many of the papers are probably available from professors' websites nonetheless.
I, for one, will start with the paper that has the best title: Churchland, Ramachandran, and Sejnowski* A Critique of Pure Vision.
[*] Screw you, UCSD, and your "standards."
Posted by todd at 5:19 PM | Comments (0)
July 13, 2005
On Deck for Connectionists Models
As if I needed more reasons to hope for admittance to UCSD. Here's a well written piece by Dr. Jeff Elman titled Connectionist Models of cognitive development: where next?. It appeared in TRENDS in Cognitive Science in March, and is a review of important and recent literature surrounding connectionist (those using neural networks) models of cognitive development. There is an emphasis on those models which deal with language acquisition.
I'm always telling people that computational models are good for directing research in the wet sciences, but I don't have a lot of concrete examples. This paper is full of them.
In one particularly cool bit, Elman discusses research which is attempting to discover how we learn to differentiate spoken words. Contrary to the way it seems to a practiced native speaker, there are no clear breaks in spoken language. Some computational models have apparently converged on one way in which this discrimination could take place:
... word boundaries are locations where the conditional probability of the next sound, given what has preceded it, is low. A network (or child) that attempts to anticipate what it (she) will hear next will tend to do worse at the onsets of words, and better as more of a word is heard. Error maxima thus constitute likely word boundaries.Neat stuff. The paper has 74 references, and looks like a good place to start for anyone interested in recent advancements in this area. (Which, given the audience here, is basically me. Still.)
Posted by todd at 11:27 PM | Comments (2)
June 30, 2005
Brain acts 'like a biological organism'!!1!
An article in the Proceedings of the National Academy of Science by Cornell psycholinguist Michael Spivey got Slashdotted and mentioned in a few news sources yesterday. The first paragraph of the stories tended to sound like this:
The theory that the mind works like a computer, in a series of distinct stages, was an important steppingstone in cognitive science, but it has outlived its usefulness, concludes a new Cornell University study. Instead, the mind should be thought of more as working the way biological organisms do: as a dynamic continuum, cascading through shades of grey.The claims of the article are actually a little more modest. Basically, they are able to conclude that the brain begins processing morphemes (words) before waiting to hear them completely. They also show that a particular mathematical model of the decision process, using dynamical systems and attractors, accurately accounts for the data obtained.
What I don't understand is what this is supposed to say about the relationship between brains and computers. At the very lowest level, my impression was that we were already sure there were no 'bits' and no clock forcing the brain to perform operations in 'cycles', as in a CPU. At higher levels, there's really nothing tying computers to representations which utilize discrete states.
The fact that the data fit the dynamical system so nicely is very interesting. Which is to say, I'm slowly building a stack of reasons to be interested in dynamical systems. I've checked this book on dynamical systems out of the library twice now, and only read about a chapter. But the source of my interest has been artificial intelligence -- a professor convinced me that to do anything interesting in neural nets I would want to know dynamical systems. So the applicability of these mathematics to cognitive science isn't news.
The news is that language processing specifically does not involve a process whereby some system in the brain discriminates words and releases the results to other systems only when a complete decision is made. This is a pretty cool contribution to an understanding of language and cognition, but you wouldn't know it from reading the news articles. You wouldn't even really understand the conclusion.
Unfortunately, the headlines which the paper generated do little but reinforce silly misunderstandings. The best way to think of a brain is still as some sort of computer. The example at hand is the process by way of which the brain takes input from the ears, and through some process ends up with a representation which causes the subject to press a button. No one seriously believes it's a digital computer, or uses a von Neumann architecture. It's not an interesting statement to say that a brain acts 'like a biological organism', and it's not meaningful to oppose biological organisms to computers, because the relevant notion of computation is much broader than that which applies specifically to the Dell on your desktop.
Posted by todd at 9:43 AM | Comments (5)