Tuesday, June 27, 2006

An MRI scanner darkly

I've got a feature in the 22 June issue of Nature about some neuroscientists useing Richard Linklater's films to probe how the brain responds differently to realistic and meaningful, but obviously unreal stimuli.

Raymond Mar and his colleagues used Linklater's 2001 film 'Waking Life', an movie made using a technique called rotoscoping, which involves turning video footage into animation. The researchers sat their subjects in a brain scanner and showed them clips of the original video footage, and then the animated version of the same shot.

They were particularly interested in the brain areas involved in attributing motives to others, and trying to work out what they are thinking and planning on doing. We will attribute intentionality to almost anything - characters in books, cartoons, a malfunctioning computer. But it seems, Mar's team found, that the brain areas involved fire more strongly when the stimuli are more realistic. But no one is quite sure why, or what this means.

The piece also looks at how our brains cope with the reality/fiction divide more generally, and the psychology of narrative. It seems the the suspension of disbelief is a myth, and that, as long as information comes in the form of a story, we are ready to believe almost anything.

One of the reasons I wrote this now is that Linklater has another rotoscoped film out next month, A Scanner Darkly, based on the Philip K. Dick novel. Dick, of course, was well into the untrustworthiness of our brain's picture of reality.

Wednesday, June 14, 2006

100 conservation questions

Everyone loves a list, including ecologists. And some of the bigwigs of British conservation have now released one with the snappy title of " 100 ecological questions of high policy relevance in the UK".

The list is grouped by topic rather than importance, which will make the five-hour special on Channel 4 less interesting. There are 14 categories, such as forestry, urban development, and invasive species. Questions include: 21. Why have many woodland birds declined? 32. What are the impacts of recreational activities on biodiversity?, 67. How can soil carbon be retained and further carbon be sequestered in the soil?

The authors acknowledge that these are rather vague. This, they say, is the product of input from policymakers, because policy is typically focussed on quite general questions. The challenge for researchers is to derive meaningful specific research projects to address these (I wonder if framing grant applications in terms of this list will help people get funding) and, conversely, to stress what can be generalized from their own tightly focussed projects.

One hundred questions seems to me rather a lot — a shorter list would have been easier to get one's head around, and so perhaps more galvanizing. But these 100 were boiled down from a longlist of 1,003 (!), so perhaps we should be grateful.

The lead author is Bill Sutherland, of the University of East Anglia, who is working hard to promote what (with reference to medicine) he calls ' evidence-based conservation': i.e., making sure that conservation practices are tested, and that they work, rather than doing things just because they're traditional, or seem like a good idea.

You'd be surprised how little we know about whether conservation strategies work or not. I touched upon this earlier this year in a Nature piece on farmland biodiversity called "How green was my subsidy?" [subscription required].

Billions of euros of EU money are spent each year on agri-environment schemes, but many have no clear goals and are not monitored. When people look at whether they do enhance biodiversity, about half seem to have no effect. It would be a mistake to be down on these subsidies as a whole, though — they take many approaches, and have many goals, and some undoubtedly do work. An international project called AE footprint is currently trying to work out how to evaluate them.

How about a global version of the same list (perhaps there already is)? My guess would be that understanding hotspots and carbon sinks and sources would feature large.

Tuesday, June 06, 2006

Sensible sheep, and cheeky monkeys

A couple of things in the most recent Animal Behaviour caught my eye.

One, Sheep self-medicate when challenged with illness-inducing foods, by Juan Villalba and colleagues, found that lambs can learn which compounds will relieve a stomach ache brought on by, say, tannins. Then, if given the same tannins in the future will choose the appropriate remedy.

I’d put this down in the cute, but not earth-shattering category. Villalba and co write that:

“From prehistoric times, people have looked to the presumed self-medicative behaviour of animals for remedies of ailments but it is still not clear whether animals seek medicinal compounds to recuperate from illness. Evidence of self-medication is based almost exclusively on observations rather than experimental analyses.”

On the other hand, there’re already whole books on animal self-medication.

The other is called Rhesus monkeys, Macaca mulatta, know what others can and cannot hear, by Laurie Santos and colleagues. The researchers set up an experiment where the monkeys competed for grapes with a human experimenter. The monkeys had a choice between picking grapes from a noisy container, and from a silent one.

If the human was looking in the other direction, the monkey preferred the silent container, presumably showing it was out to sneakily get the grape. If the human was paying attention, the monkey didn’t care which container it took — showing that the monkeys understand the connection between hearing and knowing. This also hints that they have theory-of-mind type notions of what others are thinking.