Cognitive psychology & advertising

Here’s another approach to understanding how adverts work – cognitive psychology, as discussed in this Wired article from 2002 (thanks Lauren!)

You’ll probably not be surprised that I’ve lots of sympathy for experimenal psychology as a method for understanding adverts (as opposed to, say, semiotics). A conventional experimental cognitive psychology approach to understanding something about advertising would be:

1. Have an idea, e.g., I think Factor X makes people buy more stuff
2. Come up with an experiment which involves two situations which are identical except for the presence/absence of Factor X.
3. Include some measure which is a good enouch approximation for the behaviour ‘buying’ (it could be actual purchases, or it could be something like memory for the product, or extent of positive feelings for the product, which we just assume will convert into sales)
4. Do the experiment, write up the results, let the rest of the (psychology) world criticise your experiment
5. Do follow-up experiments to re-test your idea and counter criticisms.

Or something like that anyway. Here’s an example from the Wired article:

One example: At the University of Texas at Austin, cognitive science professor Art Markman gave a group of hungry people a few bites of popcorn. Another group got no food. Then he showed his volunteers pictures of products – DVD players, shampoo, cars, French toast. The group whose appetite had been whetted with popcorn had a harder time concentrating on the nonfoods. One obvious implication, Markman says: Food samples may actually hurt nongrocery sales.

Now the strength here is that you both check if there is an effect at all, and you narrow down the possibiliies so that you have a rough idea what is causing it (again, cf a semiotic approach). The weakness is that even though you’ve shown an effect in the lab, you’re not sure it will operate outside of the lab (the problem of generalisability), and you’re not going to be sure that, even if it does operate, it isn’t made irrelevant by some other factor that you weren’t looking at with your experimental lens. So, for example, maybe wetting people’s appetites does make it harder for them to concentrate on non-foods, but maybe in real life most people don’t wet their appetites before shopping for non-foods, so the finding is irrelevant. Or maybe everyone wets their appetites, so the supermarket needn’t worry about giving away samples – we’re all peckish anyway. Or maybe we’re more likely to buy non-foods when we’re not concentrating (concentration being the thing actually measured in the experiment), so being peckish actually helps non-food sales.

Anyway, so lots of things could be true, and it takes more than a simple lab study to work out which factors are dominating, but the great virtue of the experimental method is that it gives strong hints as to what sorts of things can be operating and – just as important – what sort of things can’t affect behaviour.

3 thoughts on “Cognitive psychology & advertising”

  1. I actually ran many of the participants in the popcorn experiments. The devaluation effect is one of the coolest examples of how clueless we are about our own goals and behaviors.
    Interestingly, Art’s work has told us more than that nonfood item sales might be hurt by food samples. He and Kyungil Kim, one of his students, have found that in the morning, when people’s goals are specifically directed at breakfast foods, they actually devalue dinner foods, and in the evening, when their goals are directed at dinner foods, they devalue breakfast foods. Goals, it appears, are pretty specific. So if you’re in marketing, you’re going to want to make sure that the goals you activate are specifically those that will make people want your product (say, breakfast cereal) and not one in a closely related product class (dinner foods or snacks, say).

  2. interesting. a couple of years back i used to work for an online media network (the guys who buy and sell advertising space online) as a ad banner creative.
    i made a lot of experiments, mainly involving linguistics. it beat any lab study i’ve seen. meaning i could make two different versions of a banner. exact same graphics, timing etc, but different language. these would get posted online and by the next day, literally millions of impressions had been served and it was quite easy to see if any one approach was more successful than the others.
    somehow i doubt any university’s research department can get a million test subjects to test a theory.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: