Think of a song. A simple, well known tune. Now tap out the rhythm on your desk and ask a colleague to guess the name. Easy, right?

Well, an experiment from Elizabeth Newton, a psychologist at Stanford, suggests not. She split participants into two groups: 'tappers' and 'listeners'. The first group chose a song and then, without revealing its name, they tapped out the rhythm for the listeners to guess. The tappers estimated the probability of the song being recognised at 50%. They were wildly wrong. Of the 120 songs in the experiment only 2.5% were identified correctly.

States of mind

So what causes the gap between prediction and reality? Well, when the tapper beats out their tune they can't help but hear the song play through their head. However, all the listener hears, in the words of the psychologist Chip Heath, "is a bunch of disconnected taps, like a kind of bizarre Morse Code".

The tapper is suffering from the 'curse of knowledge' – the difficulty of imagining what it's like not to know something that we know. It's hard to recreate others' state of mind. This is problem for brands as it leads to inappropriate communications.

Think like a customer

Consider posters. Many posters are approved after poring over mock-ups in a meeting room, a scenario far removed from that of the consumer. This process results in posters being legible when scrutinised, but not when viewed from afar by distracted pedestrians.

Ifan Batey, a researcher at ZenithOptimedia, conducted a small, albeit unscientific, experiment to quantify the problem. He walked around the West End and categorised the legibility of posters from the other side of the street. He found that 4% were illegible and for more than a third, only the headline was easily read. This represents a significant waste of money.

The solution is to recognise that we struggle to empathise and instead change the context of evaluations so that it mimics the customer's experience. This can be achieved either by pre-testing copy on an actual billboard or monitoring consumers' response using excellent mock-up tools like Posterscope and JCDecaux's 'Virtuocity'.

Illegible copy is just one example of a broader problem. Marketers' experiences are unrepresentative because they spend far more time thinking about the brand than their customers. It's crucial, therefore, that they don't assume that consumer reactions to their messaging can be predicted just through reflection.

Instead, the main lesson from Newton's research is that we need to spend more time bringing a consumer perspective to bear on the message and media at the start of the planning process. Otherwise our communications may be as confusing as a bizarre form of Morse code.