Wednesday, 31 March 2010

Cognitive versus neuro models

This article (from Jonah Lerner at Science Blogs, via Farnam Street) looks quite important at first:
A recent study led by Brian Knutson of Stanford, Drazen Prelec of MIT and George Loewenstein at Carnegie Mellon [LC: a good start - I met Prelec at a seminar last year, very smart guy - and Loewenstein wrote my favourite behavioural economics book]
...when subjects were shown pictures of an object they wanted...brain areas associated with anticipated rewards, such as the nucleus accumbens, exhibited a spike in activity...
When the experimental subjects were exposed to the cost of the product, their insula and prefrontal cortex were activated. The insula secretes aversive feelings, and is triggered by things like nicotine withdrawal and pictures of people in pain. In general, we try to avoid anything that makes our insula excited. Apparently, this includes spending money...
...this data directly contradicts the rational models of microeconomics. Consumers... don't... perform an explicit cost-benefit analysis. Instead, we outsource much of this calculation to our emotional brain, and rely on relative amounts of pleasure versus pain to tell us what to purchase.
Now you can interpret this experiment in two ways. One is to assume that we need to properly understand the structure of the brain in order to understand behaviour. If this brain area is stimulated by anticipated pleasure, and that area responds to pain and prices, then it's all about neuromarketing. We need fMRI to understand our customers.

The other way is to say: never mind the brain. Cognitive models deliberately ignore the structure of the brain, and instead look at higher-level concepts like attention, knowledge and preferences. The fact that some of these preferences originate in specific brain areas is completely irrelevant.

I have to say, unless you're planning to carry out surgery on your customers' brains, the second approach is really the only one to go for. The neural input is interesting in this experiment, because it provides some evidence for what the underlying preferences might be - but in this case it tells us little that we couldn't already infer from behavioural data. Or from two hundred years of marketing experience.

Anyway, the article's still worth reading for some tips on how to lay out your store to point your consumers' attention to the preferences that suit you. The result?
...all those details of the Costco shopping experience make us more likely to spend money. The bare bones warehouse aesthetic, the discounted house brand, the constant reassurance that we're paying "wholesale" prices - it's all an effective means of convincing us to not worry so much about the price tag. As a result, we're able to focus entirely on our anticipated pleasures, which is why I walk out of the store with all this stuff I don't need.

1 comment:

M. said...

I completely agree with you, again. Till now, neuroeconomics just confirmed ideas derived from outside. What I find most annoying is that these results are usually presented as if they were objective! Reality and nothing more ...

Cheers

M.