Last weekend I read an article in the Guardian about the recent successful appeal of Amanda Knox. The article made me think about the way we do user research.
The article describes how the Italian police and the press based their case against Amanda Knoxalmost exclusively on her behaviour rather than sound evidence. The author Ian Leslie talks about how we interpret people’s facial expressions and behaviour and assume that we can understand their true feelings just from a few external cues. He points out how an encounter between two people is affected by a cognitive asymmetry that makes people believe that their counterpart’s expressions reveal their true emotions like an open book, while they themselves are able to conceal their true feelings if they like. This asymmetry gets even bigger if it’s not a face to face encounter, but for example press coverage of a person you don’t know.
I think it is important to keep this in mind when we observe users. One pillar of a user-centred design is not to ask users what they want but rather observe what they do in order to find out what they really need.Observation of behaviour is therefore a cornerstone of most methods, be it ethnographic research, contextual inquiries or usability testing. The assumption is that observed behaviour is unfiltered by interpretation by the person observed and therefore reveals “the truth”.
To a certain extent this is true of course. Observable behaviour is empirical and therefore “true” in that sense, but the motivation behind the observed behaviour and the emotions involved may not be readily accessible through observation. For example in a usability test a researcher may completely miss that a user is getting really frustrated just because that user is quite introvert or too polite to say. And when a user displays a positive or negative emotion, we should not jump to conclusions on the cause, just because we have seen a similar emotion before.
To me this emphasises how important it is that we encourage users to think aloud as much as possible as they are being observed. As researchers we need to be extremely alert and probe exactly what is going on in the user’s mind and not base our insight on our observation alone. This is especially important if any recorded session data is shown to designers or developers who are missing all the first hand information of a face to face encounter.
I think there is a risk that as trained user experience professionals we may think that we can interpret people’s behaviour and feelings more accurately than we actually can. Reading this article has made me more aware how important it is to really listen what users have to say in addition to observing them.
What do you think? I’d be very interested to hear your thoughts based on your own experience of user research and design!