Simulating the Less-Than-Perfect User

Maybe it is my age, but the assumptions that I hear made in design meetings about users and their activities completely baffle me. Team members will happily while away the time exchanging war stories about “stupid” things users do, but when they return to the design of the system in hand, it is always for the perfect user. It never seems to occur to many developers that these supposedly “stupid” things are done by perfectly typical users under normal conditions of use.

Most software and many web sites are still designed for some fictional perfect user, as shown in the illustration. What we need is some means of convincing developers that they should not rely on their own naïve concepts of user capabilities.

Insist that developers attend usability tests of their work or view video highlights. This may well be the most effective approach, but putting developers in the same room as failing users can have some unpleasant side-effects (“No! Not like that!” has been heard on more than one occasion.)
Send developers on “cognitive awareness” courses such as those run by Dr Tom Hewitt at various venues. (I reviewed his seminar at UIE’s 2001 Boston conference this time last year.) These courses may convince some of the fallibility of human cognition, but “hard core” developers may still fail to see the connection between these issues and the usability of their designs.
Attempt to provide realistic simulations of average users.

It is this last point that I would like to explore a bit further. Work of this type has been successfully done in certain specialist fields. For example, the Third Age Suit (http://www.lboro.ac.uk/taurus/simulation.htm) helped the designers of the Ford Focus to experience first hand some of the problems old age brings in the use of cars. The suit deliberately restricts some movements and makes others noticeably more difficult. Gloves and modified glasses simulate a reduction in tactile and visual perception. The overall effect is far more persuasive than the other methods in our list.

The difficulties that Ford faced are similar to ours. Automotive designers are typically young males, with no real appreciation of the difficulties that can occur in other parts of the population. Software to simulate the experience of being a real user (rather than a product’s designer) may be just as effective as the Third Age Suit. Here are some of the features SimUser might have:

Mouse randomization. Mouse clicks would occasionally occur for no reason. Some user-generated mouse clicks would be ignored while others would be displaced in space and time.
Visual fogging. Text below a particular size would be deliberately garbled. Other text would be reduced in size according to the target age group of the simulation.
Artificial disorientation. Menu items would be randomly rearranged to simulate the stress that inexperienced users face.
Lexical filtering. Words that are unfamiliar in the problem domain or to the target user communities would be garbled.
Cognitive loading. SimUser would display popup windows asking the user to perform a variety of tasks before they could continue using the application or web site. The amount of cognitive loading could be varied according to the anticipated environment.

There are other features that could be explored, such as removing some text and graphics from the screen since we know that users do not attend to all they see. However, I think the above list is a good starting point. It is no substitution for usability testing, but anything that helps developers understand the real needs of users has to be worth trying!

The Author

William Hudson is principal consultant for Syntagm Ltd, based near Oxford in the UK. His experience ranges from firmware to desktop applications, but he started by writing interactive software in the early 1970's. For the past ten years his focus has been user interface design, object-oriented design and HCI.

© 2001-2005 ACM. This is the author's version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in SIGCHI Bulletin, {Volume 34, March-April 2002} http://doi.acm.org/10.1145/967260.967270