Monday, June 30, 2008

The Large Hadron Collider

I don't know nearly enough physics to say anything useful about what is going on in particle physics. I do think that the Large Hadron Collider is spectacular, and beautiful and inspiring. The comparison with cathedrals is obvious, and I guess pretty regularly made (the words large hadron collider cathedral get about 11,000 Google hits). Really, though, this knocks old cathedrals into a cocked hat when you consider the extraordinary intellectual division of labour, and scientific and technical innovation demanded on so many fronts to pull it off. Don't get me wrong, I think some cathedrals are both beautiful and remarkable technical achievements, and understand that truly great minds worked on some of them. The LHC is just way, way, cooler.

I also have to say that if some of the doom-sayers are correct, and it might end the world, that's OK too. I don't think they are. Not being a physicist, on this point I defer to physicists, and they mostly seem to think it's either very, very, very unlikely or impossible that turning on the LCH properly will set of a large catastrophe. (See a summary of the fears and responses here.) I reckon that even if there were credible risks that we'd end the world, we should still give it a crack. The xkcd cartoon I've reproduced here suggests part of the spirit. And it would be a pretty cool way to go. Just in case, I hope to make sure I'm at least mildly drunk when they do finally turn it on. There's no desktop widget or counter that I've been able to find letting you know when the current expected date of firing up is, but you can watch the progress of the super-cooling on the main LHC website.

There's an official gallery of pictures at CERN here.

And here's a slightly old (from, I think, when they thought it would go live in November 2007) YouTube video on it:

Sunday, June 29, 2008

The Coghill Challenge

Since a piece by Ben Goldacre (also picked up by PZ Myers) got me onto Roger Coghill, I've been sniffing around a bit. I found his web store, and fired off a query about the 'research' they refer to in the case of the impressively named 'Coghill Supermagnet'.

Then I figured to see what else there was on the tubes about the Supermagnet, and soon found the website of The Medcross Group. On that site you can find the "Coghill Challenge". More on this anon.

For readers who're unfamiliar with Coghill, he's big on magnets, and electromagnetic radiation. According to him they're very bad (which is how he got to be called on by the UK Daily Mail when they wanted someone to say that cell phone towers made people suicidal) and also very good (at least when he's selling them for therapy).

So what's the challenge? You can read the official description with all it's clunky expression here. The bottom line is this (quoted with odd line breaks, etc., as in the original):

Place any human infant of less than three months age to sleep each night for at least eight hours in an ELF electric field of 100 Volts per metre for thirty days. My studies predict that child will die, or become so seriously ill that the test will have to be called off
. The NRPB and the power utilities' investigation levels by contrast predict there will be no adverse effect.

I will personally bet any NRPB member of staff or any any electric power utility worker around the world £2000 (or US$3000) willing to do this experiment, that my prediction will prove correct.

Pause. Breathe.

Yes, that's right. This guy is really confident that something would kill or harm a human infant, and his idea of a good way of making the point is to bet with real money that you can't prove him wrong. According to what he says he knows, he's bribing you to kill babies.

So, no trials with ... er ... mice (for example), then?

I'm sort of stunned. Not only do these pseudoscience bozos know very little about science epistemically speaking, it seems as though their idea of research ethics is at best about 5 decades out of date. I've emailed Coghill to ask wassup with this, and will follow up here in due course.

The "Coghill Supermagnet"

So there's a quack called "Dr Roger Coghill" who gets a fair amount of media attention in the UK, and some well-deserved smacking from time-to-time at the hands of Ben Goldacre. Goldacre recently took Coghill to task for some nonsense in the UK Daily Mail regarding a supposed link between cell-phone towers and suicide rates. (This smacking was picked up by PZ Myers and is noted in Pharyngula.)

Good quacks need a range of products, and Coghill is no exception. One that caught my eye checking out the on-line store was The Coghill Supermagnet. It's not cheap at GBP38 and some change. Here's the blurb (emphasis added):
Alternative treatment for aches, sprains, muscular pain and many other conditions.
Designed after a thorough review of the literature and intensive laboratory research, this set of magnets enables you to apply the right strength in exactly the right place and for the correct number of applications as uncovered by scientific studies from many countries.
Our unique method of superposition can deliver fields from 450 to 2450 gauss at the pole face, just in the therapeutic range. These magnets are a neodymium, boron iron mix, and are extremely difficult to degauss. In fact we guarantee them for 30 years in normal use, and they carry a 60 day money back guarantee: you simply cannot lose in trying them for aches sprains and muscular pain, but for a complete list of its many uses see our book, The Book of Magnet Healing listed in the catalogue.
So I reckon this is all bollocks. But just in case, I sent the on-line store a polite email saying I'd appreciate it if they could "direct me to the peer-reviewed research publications in which the laboratory research is reported". Updates to follow as events unfold. I'm especially curious to get to the bottom of the "unique method of superposition", and their manufacturing process.

Saturday, June 28, 2008

Durable effects of concentrated disadvantage on verbal ability among African-American children

This paper in PNAS opens noting that there's broad consensus that "cognitive ability predicts adult well-being across a variety of dimensions, whether educational attainment, labor market success, avoidance of criminal behavior, or health. There is also increasing agreement that cognitive ability is significantly shaped early in the life course, such that a focus on children is essential" (p 845).

There's also some controversy about the precise extent to which the course of early development determines cognitive ability, with some researchers (for example Herrnstein and Murray in The Bell Curve) arguing that differences in adult intelligence between populations are to a significant extent genetically based. I'm not going to try to say anything about the wider issues here, but this paper by Sampson, Sharkey and Raudenbush is an important contribution.

They key idea behind the research is an attempt to study the neighbourhood as a determinant of cognitive ability. There are several good reasons for this, reviewed in the paper, including that neighbourhood poverty is associated with inconsistent parenting, and poor mental health in parents. There are also ways in which marginal and violent neighbourhoods might be expected to compromise quality of speech communication, and in other ways limit opportunities for learning. The authors therefore hypothesised that "residing in a severely disadvantaged neighborhood cumulatively impedes the development of academically relevant verbal ability in children" (p 846).

There are non-trivial methodological problems in attempting to measure effects here. Some relate to analysis. Controlling for income, for example, while sensible in some ways amounts to hypothesising that neighbourhood doesn't itself have an effect on income, or at least reduces the chance of detecting multiple determinants of different cognitive ability. The present study attempts to deal with this in ways I’ll get to in a moment. There are also problems with selection of subjects, since the majority of studies are cross-sectional. The authors of this study opted to compare within-individual changes as related to stability or change in neighbourhood as a way of avoiding the limitations of cross-sectional approaches.

The study followed 2000+ children, living in Chicago, between the ages of 6–12. The subjects, along with their caretakers, were followed when they moved for up to 7 years. As the authors put it:

Approximately 17% of black children not living in disadvantage in 1995 moved to a disadvantaged neighborhood some time between 1995 and 2002, and 42% of the population of black children living in disadvantaged neighborhoods in 1995 moved to a nondisadvantaged neighborhood between 1995 and 2002. We exploit these ''within-individual'' changes to estimate the causal effect of moving (p 846).

Leaving out a lot of important detail, the study proceeded as follows neighbourhoods were classified into levels of disadvantage using Census data, and subjects' trajectories classified according to a series of 'waves' giving a number of sequences (e.g. disadvantage, nondisadvantage or nondisadvantage, disadvantage). Baseline disadvantage (wave 0) was not regarded as a 'treatment' but subsequent waves were. The different trajectories were related to a measure of cognitive ability combining the Wechsler Intelligence Scale for Children vocabulary test and the Wide Range Achievement Test reading examination. Subjects were selected from nearly 35,000 households, and interviewed and assessed three times over a seven year period.

For various reasons the published analysis, concerning 'concentrated disadvantage' focuses on the African-American subjects. (Concentrated disadvantage was defined as neighbourhoods falling in the bottom quartile of a distribution factoring in six elements of the Census data.)

The most important finding, and the one noted in the title of this paper, is represented in this graphic. It compares cognitive ability scores of children who started out in neighbourhoods with concentrated disadvantage and remained there for both waves of the analysis with those of children who started in the same position but moved away from concentrated disadvantage for wave 2. Again quoting from the paper "Scores for black children who lived in such neighborhoods declined sharply relative to the average rate of growth in the sample as a whole, so that by wave 2, their verbal ability scores are well below the mean and_4 points below those of black children who do not experience the treatment" (p 851).

This is mixed news. It's lousy that it seems as though early disadvantage is so harmful. (The authors are cautious about extrapolating without additional justification from their results, but it's still lousy, and only more so if extrapolation is warranted.) On the other hand, it suggests where to focus at least some policy attention: fix the places kids grow up.

The paper is also currently available on Sampson's website here.
Sampson, R.J., Sharkey, P., Raudenbush, S.W. (2008). From the Cover: Inaugural Article: Durable effects of concentrated disadvantage on verbal ability among African-American children. Proceedings of the National Academy of Sciences, 105(3), 845-852. DOI: 10.1073/pnas.0710189104

Friday, June 27, 2008

Amphibious Robot Snakes!!!

So, some engineers have built an amphibious robot snake. When I tell some people about, my voice unsteady with excitement and joy, they ask me why the engineers built an amphibious robot snake. I do not understand why anybody would ask such a question. It seems clear to me that anybody who thought they had a shot at building an amphibious robot snake would just get right down to it. (They may have had other reasons - read about it on Geek Alerts here.)

Truly we live in a golden age.

Tuesday, June 24, 2008

Neural Ensembles in CA3 Transiently Encode Paths Forward of the Animal at a Decision Point

A terrific paper by Adam Johnson and A. David Redish in The Journal of Neuroscience from November 2007 shows that neurons in the hippocampus transiently represent paths forward of an animal when the animal is deciding between more than one path.

We've known since the late 1970s that some hippocampal cells are spatially selective. The cells came to be called 'place cells' and the spaces to which they responded their 'place fields'.

We're still finding out what exactly they're up to, and how they support the computational demands of different sorts of problem. Johnson and Redish trained rats to navigate two kinds of maze for reward, in order to ask some specific questions about what part of the hippocampus was up to at short time scales. The kind I'll focus on here is illustrated here: a 'multiple-T maze', with a sequence of choices. (The picture is figure 1 from the paper.)

Each rat also had multiple tetrode recording devices implanted into the CA3 region of his hippocampus. Using Bayesian decoding, movement of the rats through the maze was related to measured activity in the hippocampus, to identify place fields and corresponding patterns of neuronal firing. (The analysis associated places with ensembles of spikes, rather than individual cells.)

The upshot of this analysis was, partly, a representation of hippocampus activity associated with the physical space of the maze.

The really exciting part of the work was relating those two (physical space and hippocampus activity) with a third stream of data relating to the changing physical position of the rat, and the specific contingencies of the task faced by the rat. Rats are known sometimes to pause and look around when they're at decision points in mazes. In a properly designed maze where observed behaviour tells us that they're pretty good at taking the turn that gives them better reward, we have a decent idea what information they're drawing on when they eventually start moving.

Johnson and Redish observed that at the decision points the firing patterns in the hippocampus showed that place fields forward of the animal, and down the available paths, were active. This very clearly suggests that the hippocampus is involved in mediating memory functions relating to decision making.

There's a telling graphical representation of the key finding, that conceals all the analysis that went into this. (My second figure is one panel of figure 7 of the paper.) The white circle shows the physical position of the rat as it moves through the maze, and the blue the place fields arrayed to correspond to the physical space of the maze. The coloured pixels on the place field show activity as recorded by the tetrodes at the time - mostly corresponding to the space where the rat is, but clearly sweeping ahead (especially at 120-280 milliseconds) while the rat pauses at the decision point.

This is even more vivid when you see an animation, and there is one in the supplemental materials to the paper - you can also view it (this is slowed down, not actual speed) here:

This animation is a really terrific example of how very complex data can be made totally compelling when presented in a coherent way, and especially one that makes effective use of the number of dimensions in the graphic. Physical space, the position of the rat, and the activity corresponding to place fields are co-ordinated, and then updated over time so that the the fact that some cells "Transiently Encode Paths Forward of the Animal at a Decision Point" is totally clear. Bravo.

Johnson, A., Redish, A.D. (2007). Neural Ensembles in CA3 Transiently Encode Paths Forward of the Animal at a Decision Point. Journal of Neuroscience, 27(45), 12176-12189. DOI: 10.1523/JNEUROSCI.3761-07.2007

Monday, June 23, 2008

Aylett on Lint

Some readers of this blag have expressed frankly dismaying scepticism about the existence of Jeff Lint, in the face of the evidence of his involvement in major works of popular culture such as the films 'Patton' and 'Funny Girl', his scripting of an episode of 'Star Trek'(*), as well as the many books and smaller selection of comics, not to mention the records of his forays into theatre and involvement in unpopular music.

I refer such doubters to the first two chapters of the careful biography by Steve Aylett, that are freely available here. Lint himself, who may or may not be dead, has a profile on MySpace here.

(*) The episode was never filmed, partly because of its wild creativity and unfilmable special effects. Apparently the MS led Gene Roddenberry to exclaim, "This isn't prose, it's gnats in formation!"

Goodbye to George Carlin

George Carlin has died, at 71 after a long and impressive career. (There's a decent overview at Wikipedia.)

One of his last filmed skits can be seen here:

Nod to Ionian Enchantment where I first heard of this, although the news is all over.

Saturday, June 21, 2008

The Road, by Cormac McCarthy

I don't read many novels. I read masses of non-fiction for work, and normally find that if I'm inclined to read, then things I 'should' be reading crowd to the front of the queue. I mostly get around to fiction when travelling (my brain doesn't seem to handle data and statistics jammed into a crowded airliner) and on holiday. When I do read a novel the bar is pretty high - not doing this very often I demand major league literary experience, or utterly seamless and engrossing schlock.

About half of my novel reading for a year or so has been stuff by Cormac McCarthy. Earlier today I finished a second reading of The Road.

It took years for my capacity to enjoy literature to recover from the damage of a major in English Literature, and I'm not going to reverse the gains by writing a review here. It's a remarkable, extraordinarily sad and compelling read. You'll keep feeling you have to put it down because you can't bear any more, then immediately picking it up because you can't leave the story behind. The book is filled with single paragraphs of great poetic beauty. Read it.

There are plenty of reviews out there. Janet Maslin has a good one in the New York Times. Michael Chabon gets in a bunch of zeigeisty stuff, and reminds me why I'm glad to be far away from literary criticism in his take on it in the New York Review of Books.

Good/expensive vs. Bad/cheap wine - follow up

So, as I said in my entry on Marketing actions can modulate neural representations of experienced pleasantness, my (very) informal Wine Society was doing an expensive versus cheap red wine tasting event last night. We had 16 wines, half very cheap, half varying distances into the price range above merely decent wines. We had the bottles in front of us, but the contents had been carefully shuffled around, and we were set the task of sampling all 16 and recording (a) whether we thought any given one was good or bad, and (b) what we thought the actual contents were.

In the good vs. bad condition we scored one point for a correct answer and lost one for an incorrect answer. The best score was +10, and the worst was -8. (Full disclosure, I got -4.)

The best anybody did in the "what is this, really" task was 3 correct items.

It's not clear what, if anything this suggests. Having the bottles in front of us confused the cues mightily. Starting off the evening with brightly coloured cocktails and curried snacks may not have been the best preparation for our pallates. It is clear that sampling 16 dry red wines in rapid succession led to most members of the largish gathering becoming more opinionated and animated. I definitely became, temporarily, more good looking and charming.

More research is required. Billionare eccentrics with a mind to sponsor this important work should declare themselves in the comment thread. I'll likely be running our next meeting, and will report the results here.

(The cartoon is from Savage Chickens - check it out.)

Friday, June 20, 2008

150 years of natural selection, plus other events

As Olivia Judson notes in her blog, "July 1, 2008, is the 150th anniversary of the first announcement of his discovery of natural selection, the main driving force of evolution. Since 2009 is the 200th anniversary of Darwin’s birth (Feb. 12), as well as being the 150th anniversary of the publication of his masterpiece, “On the Origin of Species” (Nov. 24), the extravaganza is set to continue until the end of next year."

As PZ Myers says, get ready to party like it's 1859.

[Added 22 June: There's another piece on the anniversaries, and Darwin and Wallace, in The Guardian, by Robin McKie.]

Thursday, June 19, 2008

Atrophy speeded up

No doubt many of you have heard the rumours that a late and unpublished manuscript by Hatchjaw relating to the work of Jeff Lint has been 'uncovered' and may be going up for auction in the near future. I for one find little grounds for joy in the prospect of reading an account of Lint by the paranoid and opinionated vandal who previously added so little to serious de Selbiana. I'd rather watch a film version of "I Eat Fog" directed by James Cameron. Lint was if anything the opposite of paranoid, harbouring a pugilistic yet somnolent hostility to the little sense the world apparently made from time to time. I therefore hope very much that the rumours are false.

Marketing actions can modulate neural representations of experienced pleasantness

A paper published in PNAS in January of this year has been getting a fair bit of attention. The paper, by Plassman and colleages, is called Marketing actions can modulate neural representations of experienced pleasantness, and the bottom line result is that experimental subjects rated the very same wines as more pleasant, and had greater activity in some brain regions, when that wine was presented as being more expensive.

The subjects of the experiment had been told that they were tasting five different wines, with known prices, where in fact two of the wines were repeated, each time at a different price (one at USD5 and 45, the other at USD10 and 90).

Among the other accounts of this research in the blogosphere see Ben Goldacre, Epistemist, Neurocritic and Mindhacks.

The result isn't particularly surprising, and it seems to me that some fairly obvious variations on the task would have helped get clearer on what is going on. Also, I find some of the ways the results are presented a little unhelpful.

First, let me say a bit more about how the study was conducted. Subjects rated wines (all Cabernet Sauvingnons) without price information to establish that ratings of the wines were reasonably consistent. They were. Under scanning, with squirts of wine delivered through plastic tubes, they made a series of ratings (of either 'pleasantness' or 'intensity' on a 6 point scale) interspersed with mouth rinsing. Every wine was presented along with its 'price', and subjects held the wine in their mouth for 10 seconds before rating it. One of the repeated wines was presented half of the time at actual price of USD5 and otherwise marked up to USD45. The other repeated wine was presented at its real price of USD90, and marked down to USD10.

Subjects rated the same wine as more pleasant when more expensive, and, as hypothesised, there was more activity in their medial orbitofrontal cortext (mOFC), because other studies suggest that it is in the business of representing 'experienced pleasantness'.

So far so good. But the authors open by saying that "A basic assumption in economics is that the experienced pleasantness (EP) from consuming a good depends only on its intrinsic properties and on the state of the individual." They give a citation for this claim, but as so often the citation is to another publication endorsing the same prejudice, rather than a piece of evidence from the history of economics or utility theory. Actually, even Bentham was quite emphatic that as well as direct determinants of the utility from an experience (intensity, duration, certainty and propinquity) but also three important relational properties - fecudity (the tendency of a pleasure to be accompanied by others), purity (the tendency of a pleasure not to be followed pain) and extent (the number of persons who share a pleasure).

So price could well be an indicator of fecundity or purity (suggesting fine rather than lousy food, toadying service, well-dressed dates, etc.) even to a Benthamite. I wonder if there will ever be a day when people can manage to report results like this without feeling the need to announce a caricature or worse outright misrepresentation of the history of economics.

An earlier (2004) paper in Neuron, called Neural Correlates of Behavioral Preference for Culturally Familiar Drinks by McClure and colleagues, is clearer, and covers related ground. McClure et al found that on average subjects liked Coke more when they knew what they were drinking, than when they didn't know whether it was Pepsi or Coke. In their case the neural correlates of the marketing related preference weren't the mOFC, but a set of areas including the hippocampus and the ventro-medial PFC.

We need more experiments to try to isolate effects of scarcity (for example telling subjects that what they were tasting came from one of two classes, set up so that one was much scarcer), as well as Bentham style fecundity, purity and extent. That would also help shed light on how the different components of prior beliefs are neurally implemented and how they interact.

Two more quibbles, then a speculation.

Quibble 1: The abstract of the Plassman paper, and some remarks in the paper, suggest that this has something to do with "decision making". Indirectly it probably does, but the subjects in the experiment didn't make any decisions at all - they merely reported ratings. The subjects in the McClure et al study did make decisions forced choices between Coke and Pepsi, where there was a definite opportunity cost to each choice.

We just don't know whether subjects would have kept on buying the "UDS90 wine" over the "USD10" wine having tasted both and when spending actual money. We know that people like free wine more when they think it's expensive, and that's not the same thing at all.

Quibble 2: There's some plain sloppiness here, whether the fault of the authors or PNAS. On p1051 the text refers to figure 1D as representing data from the 'postscanning blind test' whereast the figure caption indicates that it is panel 1E that represents that data. On page 1053 it says that "Two of the three wines were administered twice, once identified by their actual retail price and once by a 900% markup (wine 1: $5 real retail price, $45 fictitious price) or a 900% reduction (wine 2: $90 real retail price, $10 fictitious price)." Yeah right, a 900% reduction of USD90 leaves you with ten dollars. There I was thinking that 10 was about 11 percent of 90.

Then the speculation. I'd really like to see a roughly analogous experiment in the case of politics. Hotelling effects drive the content of at least some policies held by opposing parties in democracies towards near-indistinguishability. But citizens typically hold strong preferences between parties. Why not have subjects rate their level of agreement with policy claims and vary whether they're told that the policies are those of a Democrat or Republican, or a Tory or Labourite, etc.? I'd bet that the framing would make a difference, but I'm curious as to whether any coherent brain data would come along with that.

Oh, and one last thing. My occasional wine society (a bunch of friends who get together to eat and drink) are doing the opposite tomorrow night - a blind tasting to see if we can pick the expensive wines out of a set with 50% of the cheapest wines we can find. I'll report back. Odds are we'll drink too much and not discover anything new. Also, sadly, we don't have a brain scanner.

Monday, June 16, 2008

License to Sin: The Liberating Role of Reporting Expectations

The paper I'm writing about here is a little old (June 2007), but it's a good start to what I expect will be a series reporting research on 'dumb people'. All will relate to the many, many ways in which psychologists of various kinds (and behavioural economists)
keep on finding ways in which people depart from optimal behaviour or show failures of rationality. (And I don't mean anything 'thick' by rationality - just consistency in pursuing any set of ends.)

This piece, by Fitzimmons and colleagues, is published in the Journal of Consumer Research, and is called License to Sin: The Liberating Role of Reporting Expectations. (It's also available here.)

The key findings are pretty simple: Asking people in advance about whether they think they are going to engage in some kind of "vice behaviour" makes it more likely that they will engage in the vice. A "vice behaviour" is defined by the authors as a behaviour about which subjects "simultaneously hold both negative explicit and positive explicit attitudes", and the sort of thing they have in mind is skipping class, watching TV instead of working, drinking more than you should. One part of the study successfully confirmed the mix of explicit negative and implicit positive associations to the vice behaviours, by comparing deliberate and response time tasks.

The study found that students asked (and asked once at the first class of a 16 week semester) how often they thought they would skip classes went on to skip classes more often than students who were not asked.

It also found that subjects who already thought of themselves as having a problem with self-control were more likely to exhibit increased vice behaviour following being asked about expected vice.

Finally, it was found that the effect of reporting expectations could be reduced if subjects "precommitted" to self-reward for avoiding the vice, or thought about actions they could take when faced with specific temptations. The "precommittment" simply involved describing a reward that the subject could give her or himself if they met their vice-restraining goal (this part of the experiment concerned drinking).

None of this 'should' make any difference. Answering a question about how often I expect to muck around blagging instead of grading papers or doing research doesn't give me any new information, and doesn't seem like the sort of thing that should change my preferences. And since talk is cheap, imagining that I could pay myself for not giving into temptations seems like no reason at all not to take the temptations and then take the reward anyway. But the results suggest that both of these sorts of things do make a difference.

One reason not to pay myself the reward when I didn't 'earn' it, is, of course, because my own credibility is worth something to me. Some of my total expected reward will depend partly on that credibility. I'd have been interested to see whether thinking about ways of dealing with temptation, or planning self-reward, worked the same, or better, or worse, for subjects who regarded themselves as having control problems, or who were more impulsive by some other measure.

I also figure that the next time I throw a party, I'll put a little quiz on the invitations, asking "Do you think you're likely to drink too much and dance on the tables?", "Do you think you're likely to get frisky with the host?" and so forth. It may be harder to work this into dating without the pre-date conversation seeming a little weird, but it looks like that could also be worth a try.

Saturday, June 14, 2008

Reward prediction, yes. Stimulus categorization, no.

Here's a follow up to an earlier posting, my first stab at blagging a piece of proper (peer-reviewed) research. The paper, by Pan et al, is published in Nature Neuroscience, and is called Reward prediction based on stimulus categorization in primate lateral prefrontal cortex.

As I said before, this paper is a welcome addition to our knowledge in the field of neuroeconomics, but the results are presented in a way that I regard as partly confused or misleading.

The subjects of the experiments were monkeys. The 'reward prediction' referred to in the title is their making choices, by means of saccades to specific targets, that correctly led to greater reward while activity of lateral prefrontal cortex (LPFC) neurons was. There's nothing new about this - monkeys have been saccading to targets that predicted reward with wires stuck in their LPFCs for a while. What's relatively new is the specific challenge they faced identifying the correct target.

First they were trained that each of two different sequences of visual stimuli interspersed with saccade tasks (A1-A2-A3, and B1-B2-B3 - see figure) led to equal reward after the final saccade and stimulus. Then they were separately trained that A3 predicted greater reward than B3.

Then they were offered a 'novel' choice between A1 and B1, and 'correctly' chose A1 most of the time.

Again, as I said before, the behavioral result is cool if not surprising (it's old news in animal learning). The monkeys 'could' have showed no relative preference in the final task, given that the first stimulus in each sequence had not been directly associated with different relative reward at the end of the sequence. And the recordings from large numbers of neurons in the LPFCs of the monkeys found that some were preferentially active for rewards, some for stimuli (A1 or B1), and some for stimulus-reward interactions (e.g. associations between A1 and the differential reward associated with A3 over B3).

The title of the paper, the abstract, and various remarks in it, suggest that we should read this as having something to do with 'categorization', and that the work presents some kind of problem for temporal-difference learning approaches to reward prediction. I don't think either point is correct, even though the work is interesting.

On the first point, categorization has an established meaning in cognitive science. According to the first sentence of the entry on 'Categorization' in the MIT Encylopedia of the Cognitive Sciences, "Categorization, the process by which distinct entities are treated as equivalent, is one of the most fundamental and pervasive cognitive activities." But this paper isn't about treating distinct entities as equivalent except in the minimal sense that numerically distinct instances of A1 are treated as equivalent. If that's all it takes to make a categorization experiment, just about every neuroeconomics experiment has been a categorization experiment. (Pan and colleagues don't give a definition of categorization, so it's not clear enough what they mean by it.)

On the second point, predictor-valuation (or temporal difference) models can cope with sequences of predictors where reward only follows the final predictor. (Montague and Berns (2002) report a sequence with two stimuli, where the middle one was sometimes unpredictably timed relative to the first and the reward.) Predictors are predictors of the values of states, and associations establish informational links between states. So making a state have a higher value, which is what the middle stage of the Pan et al study did by making A3 worth more than B3, should raise the values of predictors of ways of getting to those states. This new paper sheds interesting light on how the brain handles predictor valuation in cases where predictors are chained together, but it isn't reason to think that we've found something that temporal difference learning can't deal with.

Or am I missing something?

Wednesday, June 11, 2008

Reward prediction based on 'stimulus categorization' in primate LPFC

A new paper in Nature Neuroscience adds something genuinely interesting to our growing understanding of neuroeconomics, but I've got reservations about how the authors describe the significance of their own work. The paper is called Reward prediction based on stimulus categorization in primate lateral prefrontal cortex.

The authors first trained monkeys that reward was delivered after each of two different sequences of three stimuli (see picture below) as long as the correct saccades were made in between the stimuli. (Saccadic eye movements are a common choice action in these experiments - they don't require head movements, can easily be tracked, and some important early work in the field related economic values of targets to the activity of neurons involved in making the different saccades.)

In the main experiment, following the above learning, monkeys first learned that one of the two final stimuli from the initial task was associated with a larger reward than the other, and then offered a novel choice between the first member of the two sequences. They mostly 'correctly' opted for the first stimulus of the sequence that ended with the stimulus that was now associated with relatively greater reward.

The behavioral result is cool if not surprising (it's old news in animal learning). The monkeys 'could' (according to a certain simplistic behaviourism they would) initially have showed no relative preference in the final task, given that the first stimulus in each sequence had not been directly associated with different relative reward at the end of the sequence. (The magnitudes of the rewards in the first training stage had been equal for both sequences.) And there's more - the authors recorded large numbers of neurons in the lateral prefrontal cortex (LPFC) and found that some were preferentially active for rewards, some for stimuli, and some for stimulus-reward interactions.

But the title of the paper, the abstract, and remarks dotted around in it, suggest that the authors think this has something to do with the question of whether "LPFC predicts reward by means of temporal-difference learning or by a model-based method", and to have something to do with "categorization" or "category members that have not been linked directly with any experience of reward."

This seems to suppose a conflict that isn't necessary. As Montague and Berns (2002) showed predictor-valuation (or temporal difference) models cope just fine with sequences of predictors where reward only follows the final predictor (their sequence had two stimuli with varying timing of the second stimulus and the reward). Predictors are predictors of the values of states, and associations establish informational links between states. So making a state have a higher value should raise the values of predictors of ways of getting to those states. This new paper sheds interesting light on how the brain handles predictor valuation in cases where predictors are chained together, but it isn't reason to think that we've found something that temporal difference learning can't deal with.

And 'categorization' refers to a large and complex literature in cognitive science. It would be better not to invoke it imprecisely, or as though all that was needed to make things clear was a bald contrast with "simple associative theories".

"Effortless incitement"?

When we were growing up on the farm one of the few avenues of escape we had from the soul-crushing tedium of rural life was a crate of tatty old Sci-Fi titles by Jeff Lint, including a pile of novellas and one precious number of the comic "The Caterer" (one page shown below). There was no television reception in our valley, and so we heard only garbled rumours of the animated children's entertainment 'Catty and the Major' (the Major's death scene is pictured to the left) and the nearest towns were nowhere near important enough to warrant the attentions of the rare tours of the Consolation Theatre Company. So all we had were those books and the comic. They gave us a disturbing glimpse of an exotic and glamorous world of tentacles, violence and waiters.

Lint aspired, with universally accepted success, to 'effortless incitement' in his own life, and some of his most baffling and memorable characters, especially Jack Marsden, exhibit to a very high degree the capacity to drive others, including readers, into near apoplectic fury without apparent exertion.

If anyone has an original vinyl copy of 'The Energy Draining Church Bazaar' I would be very interested to purchase it.

Harold and Kumar escape from Guantanamo Bay

I can't wait for this one if only for the terrorist-directed expectoration "Fuck you, man, doughnuts are awesome!". You can, and you know you should, check out the trailer on the official Harold and Kumar site. Part of why I'm so keen now is that I really enjoyed Harold and Kumar go to White Castle, which I thought was genuinely funny, and had some smart and still playful handling of stereotyping along the way.

Here's a fragment of dialogue from 'White Castle':

Kumar: How were Katie Holmes' tits?
Goldstein: You know the Holocaust?
Kumar: Yeah?
Goldstein: Picture the opposite of that!
Kumar: Nice!

There's also a decent comment piece at the Guardian by Peter Bradshaw, one of the folks I generally trust about movies.