.:[Double Click To][Close]:.

Saturday, October 25, 2008

New Age Experiment Goes Wrong, Hundreds Dead

What with all the fuss over the Large Hadron Collider being about to suck us all into a black hole or blow up the world or something, it's easy to forget that it's not just cutting-edge, incredibly cool physics research that can be dangerous. Even seemingly benign New Age woo can go awry and end up killing hundreds - at least judging by the results of the fascinatingly flaky Peace Intention Experiment, an update of the famous Transcendental Mediation crime reduction studies.

The Peace Intention Experiment, or as I call it the PIE (in the sky?), is the latest project from Lynn McTaggart, New Age author and alternative health guru. McTaggart's previous research has investigated whether human intention can make plants grow better and whether it can "change the structure of water". This time around, she set her sights higher - the goal was nothing less than saving lives by stopping violent conflict. But the method was very similar to that used previously: McTaggart invited her fans to think really hard about something for ten minutes or so, all at the same time (coordinated via the internet). McTaggart claims that her previous studies have already proved that the collective mind of thousands of internet users can influence events thousands of miles away, sometimes even backwards in time. Then again, it's not hard to get positive results if you perform enough statistical tests.

Anyway, this might sound harmless, but the preliminary results of the PIE have just been released and they make disturbing reading. The PIE was intended to reduce violence, but tragically, it made matters much worse. I mean obviously it did bugger all in reality, but if you go along with McTaggart's usual reasoning processes and persistently confuse correlation with causation, you'd have to conclude that it had killed hundreds. Except that Lynn McTaggart doesn't think that - we'll see why later.

Here's what happened: For ten minutes each day for eight days, 11,468 intentioneers looked at pictures of Sri Lankan people while intending - hard - that peace should come to war-torn Wanni, Sri Lanka. Specifically they intended thus -
My intention is for peace and cooperation to be restored in the Wanni region of Sri Lanka and for all war-related deaths and violence to be reduced by at least 10 per cent
It's a modest ambition - if I thought that I had the power to magically alter reality, I'd probably wish for deaths in Sri Lanka to be reduced by more than 10%. But I'm not an expert in these things. Anyway, since you've probably never heard of it, Wanni is a town in Sri Lanka which has suffered badly in the ongoing civil war between government forces and the Tamil Tiger rebels. Why Wanni, you ask? Well, McTaggart tells us that one of her main reasons for choosing it over, say, Baghdad, was that
In order to get robust experimental results, I was interested in choosing an area that the West was NOT focused on. Many areas of violence around the world are already the subject of prayer groups and intention; if we were to focus on one of them, it would be more difficult to demonstrate scientifically that the Peace Intention Experiment had a significant effect in lowering violence.
This shows some basic knowledge of the problem of confounding factors in intervention studies. Interestingly, she also seems to believe that only Western people have the power to change the world through their intentions. I bet that there are a lot of people in Sri Lanka who very much wish that their country would stop being a war zone, and they probably pray for peace - and indeed their lives - quite a lot. If I lived there I certainly would, and I'm an atheist. However, Lynn doesn't think that this is a problem. Clearly non-white people don't have the Power of Intention.

Anyway, that was the plan. What happened? Well, according to the statistician involved, before the experiment started, people were getting killed in Wanni at an average rate of 102 per week. So there was plenty of room for improvement. However, sadly, during the 8 days of the experiment, 461 people died! That's an extra 45 deaths each day. Pretty impressive - if you were trying to remotely kill civilians with some kind of psychic super-weapon. Fortunately, McTaggart et. al. have an explanation for this unexpected result:
there was a last surge of attacks by the Sri Lankan government, who wanted to quash the Tamil Tiger rebels once and for all.
So that explains the extra deaths - government forces launched a surge of attacks. People tend to die in attacks. Whether the government's decision to launch these attacks was caused by the Peace Intention Experiment is not clear, but if the human mind can magically make a plant grow to twice its normal size, surely it could make a Sri Lankan general decide to attack a town. But there's more:
In the 24 days following the experiment, violence levels immediately plunged down and stayed low ever since. The weekly average death rate in the Wanni region dropped by 49 per cent!

Interestingly enough, the Eastern part of Sri Lanka, experienced the same evolution. The death rate heightened during the experiment period, and then fell by 68.4 per cent.
Of course it's hardly surprising that deaths fell in the period following a major offensive - maybe the government was in control of the area, or maybe everyone had run away. Still, maybe McTaggart could claim credit even so - human intention works in mysterious ways after all. The death rate did fall by 49%, meaning 50 fewer deaths per week, so 171 lives were saved over the course of the 24 days after the end of the experiment. PIE worked! Unless you consider the extra 360 dead during the 8 days that the PIE was actually in progress, meaning that on balance 189 extra Sri Lankans died. Oops!

The sensible thing to do at this point would be to declare the experiment a failure and blame it on, oh I don't know, the sheep/goats effect or something. But McTaggart seems instead to be claiming that the experiment was a success, on account of the reduced violence which followed the initial terrible explosion of violence. This is known as "cherry picking". She is not perturbed by the fact that the Eastern Sri Lanka region which was not targeted by the PIE experienced an even greater reduction in violence over the same period. To me this suggests that, if anything, PIE increased violence, but McTaggart believes that it shows that PIE is so powerful that it affects whole countries even if it's only aimed at one region, like a nuclear bomb of peace. This is known as "coming up with a story to explain the data".

Naturally, McTaggart wants you to give her money to keep doing this kind of thing. After all, the PIE was only a pilot study.

So Lynn McTaggart continues to wage, and win, her ongoing battle against those who would parody her, by doing it better than they ever could. (For previous episodes see Hawk/Handsaw). But this sorry spectacle is more than just a source of cheap laughs for bored bloggers. Honestly, it is. It's actually a fascinating case study in the psychology and sociology of science. McTaggart's efforts to extract a positive result are far from unique - they are only marginally more strenuous than those of some respectable researchers.

PIE shows that if you look hard enough you can literally find any conclusion in any data set. All it takes is enough post hoc statistics and a willingness to overlook those parts of the data which don't turn out the way you'd want. The problem is that in academic science, and especially in neuroscience and psychology, there is a strong pressure to do just that. If you report that gene X is not associated with disease Y, or that brain region A does bugger all when people are thinking about B, you're going to have a harder job getting it published than if you "find something". Perhaps even more importantly, scientists can be surprisingly attached to their pet genes / brain regions / treatments and really, genuinely want to find something interesting about them. Luckily, with enough clever mathematics and creative writing up, it's possible to find something almost anywhere (details in a future post.) There's a little Lynn McTaggart inside all of us.

Is this a problem for science? Hell, yes. Can we do anything about it? Sure. I'll elaborate in a future post, but basically we need to introduce pre-registration of studies - like the current systems for the Registration of Randomized Controlled Trials, but covering pre-clinical research as well. Why hasn't this happened? Because no-one's got around to it - we're all too busy dredging our data for positive results and publishing them. Hmpf.


No comments:

Post a Comment