Thursday, 16 January 2014

Just fifteen minutes of mindfulness meditation can improve your decision making

Do you have an expensive but uncomfortable pair of shoes or jeans at the back of your cupboard that you never ever wear, but you simply cannot throw away because to do so would be to admit defeat and recognise that you wasted a lot of money? If so, you are suffering from the sunk-cost bias or fallacy.

Help is at hand in the form of a new study by researchers at INSEAD in Singapore and The Wharton School at the The University of Pennsylvania. Andrew Hafenbrack and his colleagues claim that just fifteen minutes practice at mindfulness meditation reduces people's vulnerability to the sunk-cost bias - our usual tendency to persist with lost causes because of what we've already invested.

Mindfulness meditation is all about learning to stay in the moment, and the researchers think it probably helps reduce the sunk-cost bias because the error is partly caused by memories of prior investments, and also by anticipation of regret in the future if a project or prior purchase is abandoned.

The researchers first surveyed 178 adults online and found that across the sample, a natural tendency to stay in the moment (called "mindful attention awareness") tended to correlate with being less prone to the sunk-cost bias.

In two further experiments, involving hundreds of undergrads, Hafenbrack and his colleagues found that just fifteen minutes of guided, breathing-based mindfulness meditation led to less vulnerability to the sunk-cost bias, as measured by one of two hypothetical business decisions.

Resistance to the sunk-cost bias in the first scenario required choosing to buy a superior printing press even though money had already been invested in older technology; resistance in the second scenario required making the decision to discontinue investment in a stealth plane because a rival model made it obsolete. In both cases, participants who undertook the mindfulness training demonstrated less sunk-cost bias (78 and 53 per cent resisted it, respectively, across the two studies), as compared to participants in the control conditions who were instructed to spend the same amount of time mind wandering and thinking of whatever came to mind (just 44 per cent and 29 per cent, respectively, resisted sunk-cost bias).

One further study with 156 online participants looked for mediating factors. This showed that the benefit of mindfulness meditation was mediated by less focus on the past and future, and less negative affect.

"It is particularly notable in this set of studies that increased resistance to the sunk-cost bias occurred after only a brief recorded mindfulness-meditation induction," the researchers said. Critics might wonder about the lack of a true baseline control condition. Can we know for sure that the mind wandering control condition didn't elevate sunk-cost bias? Also, we need to know more about the longevity of the mindfulness effect on sunk-cost bias - in this research the test always came right after the meditation.


Hafenbrack AC, Kinias Z, & Barsade SG (2013). Debiasing the Mind Through Meditation: Mindfulness and the Sunk-Cost Bias. Psychological science PMID: 24317419

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.


Anonymous said...

I think this is very interesting. In the book "The Science of Willpower" one of the exercises to increase willpower is to simply meditate for 5 minutes a day. And by doing this you are working your forebrain more. By working it more, you gradually increase its blood supply and density, and you tend to make decisions better because that is the part of brain which helps in decision making.

But in general, I think this type of research is awesome. Could you imagine doing body exercises to remove other fallacies we fall into, or even decreasing racisim in an objective way?

Rolf Degen said...

Like all other studies on the effects of meditation, this one does not satisfy the criteria of evidence based medicine. If you test a new medicine, you must compare it with the results of a placebo group that shares the same expectations and suggestions as the experimental group. But in meditation research, the control group is never instilled with the same expectations as the experimental group. By the way, here is a very good paper (free full text) on the subject by Daniel Simons, of the Invisible Gorilla fame:

"This failure to control for expectations is not a minor omission—it is a fundamental design flaw that potentially undermines any causal inference."

So in essence, meditation research is hyped bullshit by people who are partial to the subject. Think how you would like the stuff if it were not from a Buddhist background but from a Voodoo one.

Christian Jarrett said...

hi Rolf, you're right, controlling for expectations and motivation is important, and this study failed to do that. However, it's not clear in this case why expectations about the intervention, or greater motivation, would translate into reduced sunk-cost bias.

Rolf Degen said...

Hi Christian,

you never know in advance how expectations and suggestions influence the experimental effects. Therefore you must have an adequate control group. The authors knew this, they had a control group, but they only went half the way, not controlling for expectations. As is usual in meditation research. But even with these limitations, it does not look too good, as you may see from a report about a recent metaanalysis at the Wall Street Journal:

"The report's findings show that meditation is perhaps less effective in alleviating stress-related symptoms than is widely believed, said Allan H. Goroll, professor of medicine at Harvard Medical School-Massachusetts General Hospital, in invited commentary also published Monday in JAMA Internal Medicine. "The studies overall failed to show much benefit from meditation with regard to relief of suffering or improvement in overall health, with the important exception that mindfulness meditation provided a small but possibly meaningful degree of relief from psychological distress," he wrote."

Remember, even this small effect is artificially inflated, bcause they do not control for expectations, or, even worse, have a waiting list control, which is practically a nocebo controll group.

The trust in meditation is completely overblown, because it does not result from sobber reasoning but from a sentimental fascination with thousand years old religious practice which seem "wise" for certain minds.

Laser Tutor said...

Very helpful. Thank you.

Anonymous said...

Isn't meta analysis just a fancy word for google search?

Christian Jarrett said...


Anonymous said...

how can relax my mind before i start revising my notes??????

Post a Comment

Note: only a member of this blog may post a comment.