Let’s start with what the study was all about (quoted directly from the Science Daily article):
Memories that stick with us for a lifetime are those that fit in with a lot of other things we remember - but have a slightly weird twist.
“You have to build a memory on the scaffolding of what you already know, but then you have to violate the expectations somewhat. It has to be a little bit weird," Sederberg said.
“Sederberg has spent his career studying memory. In one of his most notable studies, he had college students wear a smartphone around their neck with an app that took random photos for a month. Later, the participants relived memories related to those photos in an fMRI scanner so that Sederberg and his colleagues could see where and how the brain stored the time and place of those memories.
Marketers will recognise that this is a close approximation of how people are exposed to advertising. Rather than specifically seeking out adverts, especially when one does not know what the next advert is that one would be exposed to; people are randomly exposed to adverts as part of their environment.
The way to create a long-lasting memory is to form an association with other memories, he said.
"If we want to be able to retrieve a memory later, you want to build a rich web. It should connect to other memories in multiple ways, so there are many ways for our mind to get back to it."
It is true that people seldom consciously try to retrieve the memory of an advertisement. However, when they are making a purchase mere exposure to a brand (or even just thinking about the brand) will unconsciously retrieve a memory of the brands advertising.
The difficulty is how to best navigate the push and pull between novelty and familiarity. Novelty tells us what is important to remember. On the other hand, familiarity tells us what we can ignore, but helps us retrieve information later, Sederberg said.Discussion:
Too much novelty, and you have no way to place it in your cognitive map, but too much familiarity and the information is similarly lost.
What that means is that context and prediction play critical roles in shaping our perception and memory. The most memorable experiences are those that arise in a familiar and stable context, yet violate some aspect of what we predict would occur in that context, he said.
My interest is in finding neuroscientific evidence that explains empirical advertising research knowledge or even just advertising wisdom. I wrote two books about this.
About two decades ago Charles Foster, now CEO of Kantar Insights Africa and Middle East then Director of MB-ADTRACK, became interested in why some multi-ad campaigns seem to work dramatically better at forming memories than single ad campaigns and also why some multi-ad campaigns fail dramatically to form such advertising memories.
Analysing the Adtrack database of ad-memory for thousands of advertisements Foster came to the conclusion that:
- When the individual ads in a multi-ad campaign are too different from each other they fail to form memories.
- When the individual ads in a multi-ad campaign are similar but different there are two things that happen:
a. If all the individual ads are launched at the same time they are unlikely to form a memory. This would be a failure of memory for the individual ads as well as the whole campaign. Often this is as a result of the marketer being so proud of his creation that he wants everybody to see the whole production. (I.e. he overestimates the public’s interest in his advertising).
b. However, if only one ad is launched, with the type of media pressure one would normally put behind an ad at launch, and then the others are brought into the flighting schedule, the result is magnificently better than what one would expect from a single-ad campaign.
This is dramatic empirical evidence of Sederburg’s theories: Too much novelty, either by the individual ads being too different from each other or by launching them at the same time is not good. However, if one ad is used to establish the required memory structures and then the follow up ads use this structure, but using a novelty twist, then the resultant memory formation is way above what one would expect.
Foster recommended that the responsible (scientific marketing) approach would be to launch one ad in a multi-ad campaign and then track how its memory is settling in among the target population. Once it has reached satisfactory levels the other ads in the campaign can then be launched.
The good news for marketers is that it does not cost 4 times as much to make 4 ads in a campaign as what 1 ad would cost. Once the central theme is established and produced the cost of the ‘surprise’ in the other three is minimal. In fact, one often has spare footage from the production of the first ad.
It should be the advertising agency’s responsibility to look for opportunities to make multiple ads before they start production, and even after they have produced the ad they should also scan the remaining footage for opportunities to make multiple ads.
On-line advertising will increasingly rely on the campaign being so structured that the basic memory structure for the campaign is created and then the on-line versions should build on this. It is especially the ‘surprise’ that will keep the audience looking at the ad to see whether it has new twist.