Your mind is leaking! And what this means for user-centered design.

I recently submitted a presentation proposal to the South by Southwest (SXSW) conference for the 2011 conference. The conference runs parallel mini-conferences on Interactive, Music, and Film. The submission was only allowed to be 1500 characters, and have an 8 word title. This blog post is a somewhat expanded version of what I submitted. Once the conference opens up voting for the presentations (presentation selection for the conference is largely based on voting by the public), I'll post some more information on the topic, and ask (beg, etc) for anyone who reads it to vote for me.

User-centered design is broken! Or at least it needs some polish.

User experience practice is based on using valid theories of how people interact with the world and think (theories of cognition) to design useful, usable, and desirable products and services.

Although cognitive science theory has moved forward since the first time these cognitive theories were applied to the practice of usability related work, newer views on cognition are currently underrepresented in the practical methods that make up user-centered design (UCD).

The Information Processing Model view of cognition basically says that the brain is like a computer, and therefore people take in information from the external world, compute (rationalize), and then spit out responses, behavior and so on. The picture below is a famous (in the HCI academic world) representation of this idea.

Cognitive science research and theories that go by names such as distributed or extended cognition argue that there are major issues with this "brain as a computer" view. Instead, distributed cognition and extended mind views claim that the way we interact with the world and think is through the selection, creation, and use of ephemeral (cognitive) systems made up of internal (brain) and external resources. These interactions of internal and external resources, which are for the most part invisible, in many cases can be viewed as the "cognitive" resources themselves; i.e. many of the artifacts around us actually become part of the machinery of thought. Whether we're talking about a pencil, spoken language, or Google searches, these external technologies are just as much a part of the cognitive process as neurons. As Andy Clark says, "Mindfulness is just matter, nicely orchestrated." The human brain is still of course an integral part of the mind, but simply put the mind does not equal the brain.

So what's the big problem?

The problem is these newer views on cognition are currently underrepresented in the practical methods that make up user-centered design, or in other words, user-centered design practitioners are likely only using outdated views of cognition. As shown in the diagram below, current use of ideas from distributed cognition and the extended mind is primarily by academics, theorists, etc, and not by your everyday user experience practitioner.

How does impact day to day UX practice?

Increase evaluation and design attention on the cognitive role of the external world, but beware of noticing only the high-tech as relevant. Because many of the most useful and usable artifacts simply become invisible over time (and people then get credit for successful use), it is important to proactively pay greater attention to the possible role that multiple external resources are playing. Therefore, practice should move towards using representations that highlight interactions over time and space between multiple resources. For example, service blueprints, used in service design are a good role model for most work in interaction design. Popular representations in usability work such as wireframes that focus more on static views of interactions need to be complemented with better representations of the dynamic aspects of cognitive accomplishment. We need to beware of the trap of noticing only the most high-tech artifacts in the situation, and subsequently failing to continue to look for other more everyday resources that may be contributing to accomplishment.

For example, the "context of use" is a phrase heard a lot in user-centered design. Usually, the context of use is taken to mean the environment (physical, social, and cultural) that you are designing for; user centered designers have borrowed from the field of anthropology and now conduct what are called contextual interviews, where you interview a potential user in the actual environment where your product will be used.

This is a good example of how to cross the bridge to using ideas from these new distributed cognitive views; we can now extend how we view the context of use. In the original cognitive view (where the user carries out all thinking), the context is full of items such as paper forms, post-it notes, etc that the user may use while also using the more obvious technology such as computers, software, etc. With a newer, distributed cognitive view, we would want to understand how these simple items might really be viewed as being part of the thought process of the user. For example, if in your user research, you see many monitors sprinkled with post-it notes that are used to keep track of information while software is used, you can be pretty sure this represents a missing feature in the software. Don't ignore this...try to see this as a need that the user has, but one that the user may never tell you should be part of the software. Whatever the post-it note is helping the user remember or accomplish, should probably become part of the future design.

This leads into a second recommendation for everyday UX practice...don't assume users know things such as how they do what they do, how they know what they know, so don't expect them to be able to tell you.

 Don't base your research and designs on what users tell you. Extended cognition shows us that not all of cognitive processing that takes place is known to the individual person. Therefore, we should not expect users to be able to articulate how they accomplish things; the risk is that when asked, people will make up realistic explanations regardless, making it difficult to distinguish between the known and the made up. The simple principle here is to learn what people do and need by observing them, not only by asking...you will get an answer if you ask, and it's hard to tell the difference between what someone thinks they do, and what they do...if you don't observe. They won't consider the post-it note as part of the "system," so they won't talk about it.