TiChi Introduction

Fitt's Law
Object-Action Interface
Prescriptive Theories
Fisheye strategy
Conceptual, semantic,
    syntactic, & lexical

Direct Manipulation

Information Processing
Hacker's Action Theory
Attention & Memory
Andersen's ACT-R
Knowledge & Mental
Social & Cultural

Theories in Computer human interaction
Printer Friendly Version
Prescriptive Theories
Kendra Knudtzon
October 2002


Prescriptive theories are typically guidelines or rules to be followed in design. Rather than addressing a particular problem, or modeling behaviors, prescriptive theories offer high-level guidance to designers.

There are many books and websites dedicated to giving principles, guidelines, or heuristics to follow, for interface and website design. In a growing field, such as HCI, prescriptive type theories are often some of the first developed, because these guidelines tend to arise from experience, observation, or noticing commonly made errors. This of course, makes prescriptions very qualitative in nature.

Side note: Normally a theory is something that is verifiable and falsifiable. Prescriptive theories are neither verifiable nor falsifiable; but in the context of HCI research it is helpful to refer to these prescriptions as a theory, in part because it stresses their importance in this growing field. Especially at this time, as the computing paradigm is shifting from a technology–centric to a human-centric approach (as is presented in Leonardo’s Laptop and The Invisible Computer), the interface becomes the key ingredient in allowing people to focus on their actions, rather than the technology they use. Prescriptive theories aim to help guide designers create human-centric systems.

Scope, Application, and Limitations

Because of their high-level nature, prescriptive theories have fairly general scope.  By looking at items like known dangers or best practices, prescriptive theories tend to be very general and hard to categorize. Generality also leads to the problem that while these theories give guidelines for system design, how to actually follow the guidelines is left up to the designer. The goal then is to try to make the guidelines clear, and when possible, explain (perhaps for particular contexts) how to implement the guidelines.  Good prescriptive theories should be complete, and they should be informative of what aspects need to be considered. Theories should not be contradictory. One possible future direction is to create tools that could help designers apply these principles successfully (as Hamilton suggests in his article, “From theory to practice: Tool support for task-related principles" http://www.upb.de/fachbereich/AG/szwillus/chi99/ws/PosPap/fraser.html)

The rules and guidelines described later on this page are theories for interface design. The prescriptive theories presented here are ones that are highly referenced in HCI literature. Also included are some pointers to prescriptions for website design, which are more specific, but on the whole less useful for general interface design.  One more important thing to remember is that prescriptions or guidelines only work if people buy into them. In HCI it is important to understand why certain kinds of guidelines are important, when guidelines can be applied, and when new guidelines are needed.


In his book, Designing the User Interface, (http://www.awl.com/DTUI/) Ben Shneiderman gives his "Eight Golden Rules of Interface Design," as follows:

  1. Strive for consistency. This is a rule that is often hard to follow, because it is a rather elusive rule. There are many levels of consistency, including terminology, menus, fonts, color, and layout.
  2. Enable frequent users to use shortcuts. Interface design should include a way for experienced users to reduce the amount of time interacting with the program.
  3. Offer informative feedback System feedback is important for all user interactions.
  4. Design dialogs to yield closure . Actions should have a beginning, middle, and end.
  5. Offer error prevention and simple error handling. The goal should be to design a system where users cannot make serious errors. Errors should be detected by the system and the user should be given instructions for recovery.
  6. Permit easy reversal of actions. Users should be able to reverse actions (if possible within the context).
  7. Support internal locus of control. Users should feel as if they are in control of the software, not the other way around.
  8. Reduce short-term memory load. Information should be limited, users can only remember “seven plus or minus two” pieces of information.

In the Design of Everyday Things, Donald Norman offers four principles of good design: (http://www.jnd.org/books.html#DOET)

  1. State and action alternatives should be visible
  2. Conceptual model with a consistent system image
  3. Interface should include good mappings that reveal the relationship between stages.
  4. User should receive continuous feedback.

In a previous version of the book (titled The Psychology of Everyday Things), Norman lists 7 principles of design: (Chapter 7)

  1. Use both knowledge in the world and knowledge in the head. (conceptual models)
  2. Simplify the structure of tasks. (make visible the invisible, feedback, don’t take away control)
  3. Make things visible: bridge the gulfs of Execution and Evaluation. (show what’s possible, actions shoud match intentions)
  4. Get the mappings right. (exploit natural mappings)
  5. Exploit the power of constraints, both natural and artificial. (make it easy for user to do the right thing)
  6. Design for error. (expect errors to occur, make it easy to reverse actions)
  7. When all else fails, standarize. (if you need arbitrary mappings – standardize the system, make it consistent so that it is easy to learn.)

Jakob Nielsen lists ten usability heuristics that are supposed to be helpful for "quick, cheap, and easy evaluation of a user interface design." (http://www.useit.com/papers/heuristic/heuristic_list.html)

  1. Visibility of system status
  2. Match between system and the real world
  3. User control and freedom
  4. Consistency and standards
  5. Error prevention
  6. Recognition rather than recall
  7. Flexibility and efficiency of use
  8. Aesthetic and minimalist design
  9. Help users recognize, diagnose, and recover from errors
  10. Help and documentation

Bruce Tognazzini offers a whole discussion of principles for the design and implementation of interfaces in his article "First Principles". Many of these echo the prescriptions above, but also the article also gives more concrete advice for GUI environments or websites. http://www.asktog.com/basics/firstPrinciples.html

There is a lot of overlap among these lists: visibility, consistency, user control, feedback, error handling and reversibility of actions are all very common themes.  These items begin to address ideas of universal usability – the guidelines exist to help the developer design better and more usable interfaces. 

In addition to these lists, there are also books that deal exclusively with interface or website design. These books tend to give very detailed guidelines, mostly expanding and specifying the general guidelines presented here. Often it can be helpful to state explicitly what may be implicit in another prescriptions. The explicitness can help clarify, or really help emphasize the importance of a certain principle. There are extensive websites that offer guidelines on everything from e-commerce usability to users with disabilities to color usage to federal specifications for usability. Website design is a particularly popular topic for creating guidelines. The following websites offer links to other prescriptions, pointers to books, or general useful information:

http://www.usability.gov/ - National Cancer Institutes resources for website and user interface design

http://www.w3.org/TR/WCAG20/ - Web Content Accessibility Guidelines 2.0

http://www.usability.gov/guides/index.html - webdesign stuff

http://www-3.ibm.com/ibm/easy/eou_ext.nsf/Publish/572 - This site does a good job of explaining what guidelines are, how guidelines are to be used, and where the guidelines come from (for website guidelines)

http://usableweb.com/topics/000862-0-0.html - usableweb.com is a huge collection of information about web usability; this is a subcollection that lists guidelines and book references.

The Apple/Macintoosh and Windows guidelines for user interfaces are the standards that those corporations have outlined for developers. While the examples of prescriptive theories given above are pithy, short, understandable statements about interface design, the following websites give extremely detailed guidance to developers designing interfaces for specific platforms. These guidelines outline not only general interface concerns (similar to the ones given above), but also details about every aspect of an interface, from each type of menu, to windows, controls, icons, behavior, colors, etc; giving specific how-to instructions for conforming to their interface design.

http://developer.apple.com/techpubs/mac/HIGuidelines/HIGuidelines-2.html - The Apple/Macintosh human interface guidelines.

http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnwue/html/welcome.asp - The Microsoft Windows Guidelines for User Interface Developers


Given the nature of prescriptive theories, the principles themselves lead to or serve as examples. The prescriptions deal with a user’s interactions within a system, therefore it is difficult to demonstrate with a simple static example.  In the following (limited) example, several of the prescriptive theories are followed, but several are not, making this tool difficult to use.

NeatTools can be found at http://www.pulsar.org/2k/neattools/

Highlighting a few of the prescriptions: Dialogs do yield to closure, as can be seen above, but the system is not very consistent nor does a user feel in control of the system. The buttons along the side are the main way to interact with the system, and there are only shortcuts for a few of the commands (not shown). Options are not clear; each abbreviation needs to be learned before the system can be used efficiently. There is also no easy way to reverse actions (not shown). There is little feedback in this system, and it the conceptual mapping is not readily apparent.

 Examples such as these are encountered everyday. Unusable systems are quickly noticed due to frustration or irritation. Usable systems often can be taken for granted. As an HCI researcher, these examples and a study of prescriptive theories followed or ignored can serve as valuable design lessons.

Applicability to HCI

The prescriptive theories described above have been developed specifically for interface design for HCI. The principles have been developed from user and designer experience, and as such, are directly applicable to the HCI field.

Most people have a conception of what a good design is, and often this may rely more on aesthetics than on usability. If designers rely on aesthetics, or only on their own experiences and perceptions when designing interfaces; unusable systems often result. These prescriptive theories can help designers to consider all aspects of their interface, and what are the important areas to focus on in design and in testing an interface.

The prescriptive theories of Shneiderman, Norman, Nielsen, and Tognazzini are really technology independent. They focus on good interface design in general, and hence are applicable to traditional GUI interfaces, as well as virtual reality, three-dimensional or natural language interfaces. These prescriptive theories were based upon how the human mind works, and how it processes information, and not how current technologies work; therefore these same theories can be applicable even as our definitions of a typical interface changes.

Influence: The principles or guidelines presented in this paper are some of the seminal works of (prescriptive) theories in HCI. These theories have great influence on the HCI community: many of the websites and books linked above in the “Principles” section build upon these guidelines and many HCI courses focus on leveraging from these theories to teach students about good design.


Hamilton, Fraser et al. “From Theory to Practice : Tool support for task-related principles”. 1999. Accessed in October, 2002. http://www.upb.de/fachbereich/AG/szwillus/chi99/ws/PosPap/fraser.html

Nielsen, Jakob, Norman, Donald, and Tognazzini, Bruce. The Neilson-Norman Group website. 2002 http://www.nngroup.com

Nielsen, Jakob. "Ten Usability Heuristics" website. Accessed in October, 2002. http://www.useit.com/papers/heuristic/heuristic_list.html

Norman, Donald. Design of Everyday Things. New York: Basic Books, 2002.(http://www.jnd.org/books.html#DOET)

Norman Donald. The Psychology of Everyday Things. New York: Basic Books, 1988.(http://www.jnd.org/books.html#POET)

Norman, Donald. The Invisible Computer. Cambridge MA, MIT Press, 1998. (http://www.jnd.org/books.html#invisible)

Shneiderman, Ben.  Designing the User Interface. Reading, MA, Addison-Wesley, 1998. (http://www.awl.com/DTUI/)

Shneiderman, Ben. Leonardo’s Laptop . Cambridge, MA, MIT Press, 2002. http://mitpress.mit.edu/main/feature/leonardoslaptop/index.html

Tognazzini, Bruce. "First Principles" website. Accessed in October, 2002. http://www.asktog.com/basics/firstPrinciples.html

Talin. "A Summary of Principles for User-Interface Design" website. August, 1998. Accessed in October, 2002. http://www.sylvantech.com/~talin/projects/ui_design.html

Wallingford, Eugene. Software Systems Course Webpage. Spring Semester, 2001. http://www.cs.uni.edu/~wallingf/teaching/171/sessions/session06.html