Comments on: Do you believe in Mental Models? http://bokardo.com/archives/do-you-believe-in-mental-models/ Interface Design & UX by Joshua Porter Wed, 28 Sep 2011 20:28:57 +0000 hourly 1 http://wordpress.org/?v=3.3.2 By: Wolfgang http://bokardo.com/archives/do-you-believe-in-mental-models/#comment-1449 Wolfgang Thu, 01 Sep 2005 15:09:54 +0000 http://bokardo.com/archives/64/#comment-1449 Your discussion appears to me as a more theoretical one. While it is quite difficult to grasp the idea of a mental model and its inherent qualities and structures, the practical side is simply: Uselabs show the importance of mental models. Maybe the desciption is better coined buy: the building of a hypothesis. Any perception during the orientation phase entering a website which may lead the user to a wrong hypothesis of what this is all about can damage all IA intentions badly. I remember the uselab situation when two dynamic news items and two navigation elements concerning trainings on the homepage lead the user to the powerful hypothesis that this site is all about trainings. The many other elements (about 90% of all content elements) contradictory to this hypothesis were not any more relevant during the tasks that were to be completed afterwards. About five seconds for building a mental model of the site in comparison to ten minutes of trying to complete a given task - without further changes of the mental model. This gives some evidence about the power of a mental model, once it is formed. The IA task seems to be to assure that all content elements work together smoothly for building a more appropriate hypothesis or mental model. In this case the dynamic news items had to be reduced to maximum one for every theme or content area. But we have to admit nevertheless: It seems to be impossible to ensure appropriate mental models because they are not independent to preconceived notions, expectations, asscociative thinking etc. which is almost completely out of our reach. Your discussion appears to me as a more theoretical one. While it is quite difficult to grasp the idea of a mental model and its inherent qualities and structures, the practical side is simply: Uselabs show the importance of mental models. Maybe the desciption is better coined buy: the building of a hypothesis. Any perception during the orientation phase entering a website which may lead the user to a wrong hypothesis of what this is all about can damage all IA intentions badly.
I remember the uselab situation when two dynamic news items and two navigation elements concerning trainings on the homepage lead the user to the powerful hypothesis that this site is all about trainings. The many other elements (about 90% of all content elements) contradictory to this hypothesis were not any more relevant during the tasks that were to be completed afterwards. About five seconds for building a mental model of the site in comparison to ten minutes of trying to complete a given task – without further changes of the mental model. This gives some evidence about the power of a mental model, once it is formed.
The IA task seems to be to assure that all content elements work together smoothly for building a more appropriate hypothesis or mental model. In this case the dynamic news items had to be reduced to maximum one for every theme or content area.
But we have to admit nevertheless: It seems to be impossible to ensure appropriate mental models because they are not independent to preconceived notions, expectations, asscociative thinking etc. which is almost completely out of our reach.

]]> By: bill h-d http://bokardo.com/archives/do-you-believe-in-mental-models/#comment-404 bill h-d Sun, 03 Apr 2005 16:08:32 +0000 http://bokardo.com/archives/64/#comment-404 I don't want to beat a dead horse with my responses inspired by Activity Theory, but your question here is worth taking a shot at. Plus there is a dash of Suchman in this one... I believe that users' mental models, at the operational level which you discuss above and which techniques like GOMS render in detail, are "real" but not at all useful for design purposes. The simple reason is that these models, such as they are, are composed improvisationally...in response to situational variables that run the gamut from low-level physical conditions (like a mouse clicking) to very high-level somatic and affective conditions (like the user being jacked on too many espressos). The level at which mental models correspond to design is not the immediate operational level, but the motivational level, the level where explcit goals can be matched to end-states that are desirable. Not "boy I'd like to click three times to confirm my order" but "boy, I sure hope Josh's birthday present gets there on time." Designers would do well, in many cases, to ask not what users want to do...but rather who they want to be or become. Sounds odd, eh? Well, the *only* reasons I subject myself to the torturous, multi-level UI of my cell phone (handset, modal UI, network carrier, calling plan...) is that, in the end, it lets me "be" a better person - a more caring spouse, and here lately, a more attentive father. If Nokia could design with those goals in mind...then we might have something... I don’t want to beat a dead horse with my responses inspired by Activity Theory, but your question here is worth taking a shot at. Plus there is a dash of Suchman in this one…

I believe that users’ mental models, at the operational level which you discuss above and which techniques like GOMS render in detail, are “real” but not at all useful for design purposes. The simple reason is that these models, such as they are, are composed improvisationally…in response to situational variables that run the gamut from low-level physical conditions (like a mouse clicking) to very high-level somatic and affective conditions (like the user being jacked on too many espressos).

The level at which mental models correspond to design is not the immediate operational level, but the motivational level, the level where explcit goals can be matched to end-states that are desirable. Not “boy I’d like to click three times to confirm my order” but “boy, I sure hope Josh’s birthday present gets there on time.”

Designers would do well, in many cases, to ask not what users want to do…but rather who they want to be or become. Sounds odd, eh? Well, the *only* reasons I subject myself to the torturous, multi-level UI of my cell phone (handset, modal UI, network carrier, calling plan…) is that, in the end, it lets me “be” a better person – a more caring spouse, and here lately, a more attentive father.

If Nokia could design with those goals in mind…then we might have something…

]]>