« Google browser as nanny | Main | sIFR in perspective »
June 06, 2007
Cognitive lock-in
Cognitive lock-in: John Timmer reports on a scientific study of interface preferences... bottom line is that the more time you spend with an interface, the more you prefer it to alternatives. I think this may be because there's less time spent trying to figure it out, fewer surprises, fewer WTF moments. "Overall, the results suggest that all the years of arguments over the relative merits of things like the Mac and Windows user interfaces were a waste of time: we're generally convinced that whatever we're familiar with is the best." (Maybe it's similar to current pop music's over-reliance on pentatonic scales and 4/4 time -- easier for more people to digest than additive times or phrygian scales, even though Balkan music is really satisfying if you gain an ear for it.) If so, then the next question is "How can I make a new interface more predictable?", and a component-based interface can therefore add a lot to initial attractiveness. You have thoughts on this...?
Posted by JohnDowdell at June 6, 2007 08:35 AM
Trackback Pings
TrackBack URL for this entry:
http://weblogs.macromedia.com/mtadmin/mt-tb.cgi/8787
Comments
Nice reference to Balkan music there. I'm also tired of endless reliance of 4/4 in pop music, and can see no good reason for it. How big a hit was "Bohemian Rhapsody"?
I am a Burt Bacharach fan, one of the most commercially successful pop composers. Many of his songs include 'unexpected' harmonic and rhythmic changes, but nevertheless became massive hits. (Odd bars of 2/4 sprinkled into an otherwise 4/4 song - examples: "Alfie" and "Say a little Prayer", not to mention the 13/8 measures in "Anyone who had a heart"). Burt is quoted as saying that some of his early songs were ruined by record company execs forcing hm to remove these quirks, and dumb down to 4/4, and less of the weird jazz chords if you please... but this ended up being his inspiration to go into producing music, rather than just songwriting.
The question is not "how can I make a new interface more predictable", but "how do I make unexpected interface elements behave in a way which is both intuitive and satisfying".
I'm using the word "intuitive" in the literal sense of 'without instruction' rather than the 'warm and fuzzy' definition it usually has.
So, it's not so much about predicting the result in advance, but feeling like you just did 'the Right Thing' afterwards without any speculation about it beforehand. If prediction becomes a conscious process, something is wrong. It's not just about how the user manipulates the GUI, but just as much about how the GUI manipulates the user into making the correct choices *without* conscious prediction being involved.
Burt Bacharach and Balkan/mediterannean folk music show us that we don't have to lock ourselves into a hackneyed old pattern in order for people to feel comfortable. It takes courage, certainly, and I'm not advocating 'eye candy' as the solution either, just a preparedness to go with something which isn't always "square", (daddy-o). At least as far as a user test.
Posted by: Brennan Young at June 10, 2007 01:32 AM