Ben Goldacre, The Guardian, Saturday 9 July 2011
Since I was a teenager, whenever I have a pivotal life event coming – an exam, or an interview – I perform a ritual. I sit cross-legged on the floor, and I imagine an enormous golden beam of energy coming out of my arse.
I picture this anal beam passing through each layer beneath me, through the kitchen of the flat below, through the shop, and its basement, past gas pipes and sewers and then deep into the earth, where it spreads out into a glorious branching root network sucking power from the earth. I picture this energy surging through me, I visualise the outcome I want, in enormous detail, and I will it to happen, for about five minutes.
Surprisingly enough, this nonsense is broadly supported by data from randomised controlled trials.
One example was published last month. Around 200 students were randomly assigned to four groups, each with activities supposed to increase their fruit intake. The control group just repeated their goal to themselves (“eat more fruit”). One group concentrated on elaborate mental images of themselves enjoying fruit. Another group repeated verbal plans for specific situations (“when I see fruit, I will … “). The last group pictured lavish plans of encountering fruit, picking it up, touching it, eating it.
Among participants eating lots of fruit already, four portions a day, there wasn’t much change. Among people eating less fruit to begin with, one-and-a-half portions a day, everyone increased their intake, but the ones performing the most elaborate mental imagery did so much more (their intake doubled).
It’s not a perfect study – I don’t like subgroup analyses for a start, and it only followed up participants for seven days – but it’s not alone. An earlier study from 2009 randomly assigned 100 students either to a control group, or to a couple of forms of imagery, picturing themselves choosing a healthy snack over an unhealthy one. The imagery group had more healthy snacks.
Meanwhile, a meta-analysis from 2006 collectively analyses the results of 94 studies and finds that “implementation intentions” (“if I am in situation X, I will do Y”) had a positive effect overall on goal achievement.
So there’s probably something there, and this research tells us some interesting things about science. Firstly, I think this kind of research is useful. Rupert Sheldrake is the researcher who claims dogs can sense their owner is coming home before they arrive. I disagree with him on a lot, but he has one great idea: that each year, a proportion of the research budget – a hundredth, a thousandth – should be spent on whatever the public vote for. Most of it would go on MMR and homeopathy, of course, but some of it might go on testing, revising and improving stuff that improves people’s everyday lives.
Secondly, it shows us that even if you’re wrong about how something works, it might still work. I was sold the golden bum beam stuff with a lot of nonsense about quantum hippy energy, but I’ve always thought of it as a perfectly sensible way to combat distractibility. Effective things can come from silly places.
And so, lastly, I’m cornered into saying something nice about a government. The Cabinet Office Behavioural Insights Team annoyed me. It looked like they were going to overextrapolate from behavioural economics research to make lavish, overstated, untested claims.
In fact, they’ve just published their report on reducing energy use, and after setting out their ideas (they reckon, for example, that giving people detailed feedback and suggestions on energy use will reduce it overall) announced they’re going to test at least some of their ideas, in randomised controlled trials, before implementing them, to find out if they work. It’s odd, but the first good trials in UK politics for many years may be about to come from the wackiest and most vogueish corner of government.
[Sorry, I forgot to post this in July, when I was v busy!]