Lecture : More Human-Computer Interaction

In the previous lecture:

• What is a labelling system? When is it used?
• How do we determine effective labels for different situations?

In this lecture:

• How and when should we design things so that people immediately know how to use them?
• Why are some things easier to use than others?

References:

Norman, D. The Design of Everyday Things, Doubleday, New York, 1990 (and other books on design).

Why are so many of our artefacts so difficult to use? ...because they have been poorly designed! This is not good enough!

Objects that "work" offer a psychology of use that is intuitive and "natural".

• The steering wheel of a car
• An analog clock
• An orienteering compass
• What are some other common objects that "work"?
  

Spherical compass	Pocket compass (antique)	Gimbled Ship's Compass (antique)	Silva orienteering compass

Good design defuses the tension between functional and aesthetic goals.

Pinarello time-trial bicycle	Ferrari	Alessi juicer (by Philippe Starck)	Waterman fountain pen

Some companies (such as Ferrari, Pinarello, Waterman & Alessi) are famous for their designs.

In any real-life situation there are always competing pressures of cost, quality and production time.

Cognitive Engineering (...engineering the way people think)

• Affordances of objects.
  
  Affordances are the perceived and actual properties (mainly functional) that determine how an object can be used.
  
  Different materials have different natural affordances (that we have learnt largely through our interaction with the world).
  
  

  Balls - throwing Knobs - turning Handles - grasping

  Glass - looking through (or breaking) Paper (and toilet doors) - writing on

  
  
• Complex things require explanation, simple things do not.
  
  If simple things require pictures, labels, or instructions, the design has failed.
  
  

  Make the affordances visible.
  
  

  One estimate states that we "know about" over 30,000 discernable objects.
  
  
  
• Constraints of objects.
  
  These limit the possible ways in which functionality can be interpreted.
  
  

  Finger holes in scissors Push-plates on a door Non-square shape of 3.5" floppy diskettes

  Functions on a digital watch Operations entered on a command-line UNIX interface

  
  

  The more possible interpretations a thing has, the more difficult it will be to use.
  
  

• Mappings of objects
  

Mappings (in general) are relationships between things.

In terms of design, mapping might be between controls, their movements, and the result of the operation on/in the world.

Natural mapping leads to immediate understanding because it takes advantage of physical analogies (folk physics) and cultural standards: a close and natural relationship between control and function.

Physical Analogy
A car steering wheel - the direction the driver turns the wheel top matches the direction the car steers.   A formula-1 car steering wheel. Is this intuitive? (http://www.f1technical.net/article30.html) How does this device make affordances visible to the driver?
An aeroplane joystick - pushing it forward (away from the pilot) drops the nose of the aeroplane (away from the pilot). The joystick (front/bottom and left of center) is just one of many visible controls. A pilot does not have time to cycle through a series of menus to locate various functions of his control. (http://staff.tay.ac.uk/bstmjc/mjc_research1.htm)

Cultural standard
Rising level - indicates 'more' of something and vice versa, as does an increase in numeric value and, in Western (English speaking) culture, left to right.
Light switch position - up indicates 'off', down indicates 'on' (in Australia) What might an increase in pitch (of an audio tone) indicate?

Conceptual Models

• Conceptual models are mental models of the way an object operates.
  
  
• Many good designs provide appropriate feedback to confirm the user's mental model of operation.
  
  
• Effective use of affordances, constraints and mappings arise largely from careful consideration of visual structure to help the user build a good conceptual model.
  
  
• A user's conceptual model may not always match that of the designers.
  
  
• With a good conceptual model, the user is able to anticipate and understand the consequences of their actions.
  
  
• If something goes wrong and the user does not understand why, this probably indicates that they have an erroneous conceptual model.
  
  
• The user's conceptual model is formed through interaction with the device and the system image - an object's physical structure, including documentation, instructions and labels.

• People will always make mistakes (too err is human…)
  
  
• Bad design blames the user for incorrect operation
  
  
• Complex systems still need to be learned
  
  
• People's understanding of the world is not, in general, the same as that of ‘experts’
  
  
• Refusing to believe the evidence — instruments indicate that something is wrong — is something really wrong or are the instruments wrong?

The action cycle: execution and evaluation

Some design questions

How easily can a user...

...determine the function of the device?

...tell what actions are possible?

...tell if the system is in the desired state?

...determine mapping from intention to physical movement?

...determine mapping from system state to interpretation?

...perform the action?

...tell what state the system is in?

Knowledge

Much knowledge is "in the world" not "in the head". Most of our "in the head" knowledge is vague and imprecise. (e.g. try to remember the layout of the QWERTY keyboard).

Information a person needs to perform a task: put as much in the world as possible. Use physical and cultural constraints.

Types of knowledge:

• Declarative knowledge: knowledge of (facts and rules, e.g. stop at red lights)
  
  
• Procedural knowledge: knowledge how (process and method, e.g. catch a ball)

Knowledge in the world:

• can be retrieved whenever visible or audible
• is interpreted rather than learnt
• can be inefficient if there is a need to find and interpret information
• has a high ease of use at first encounter
• often can be unaesthetic and inelegant, thus requires the skill of a designer

Knowledge in the head:

• requires a memory search or reminder
• requires learning (Learning is easier with a good mental model)
• can be efficient
• has a low ease of use at first encounter
• is invisible (which can lead to better aesthetics)

Applications of cognitive engineering to multimedia and user-interface design

Visibility: make the functions clear. Differentiate opposing functionality. Use visual function to confirm the user's mental model of operation. Sometimes sound can be used to make things ‘visible’ (E.g. A vacuum cleaner clogging up can be audibly detected).

Feedback: give actions immediate feedback, to reinforce the user's mental model.

Design for error: people make mistakes for many reasons. Good design accepts mistakes as a normal part of operation and accommodates them.

Forcing functions: provide constraints of operation in certain situations.

Metaphors "In theory, an interface designer adopts a metaphor as a means of making the application easier to learn. This theory is based on the premise that the user will be able to transfer his or her knowledge of a familiar object structure to the new application. In practice however, an interface designer often adopts a metaphor as a means of expressing his or her model of how the system is organized. That the metaphor might make the application easier to use is often an after-the-fact and unsubstantiated rationalization of the design." (Interface Hall of Shame)

Apple Quicktime movie player interface	Omni Group Omnigraffle drawing package interface

Summary:

• Use both knowledge in the world and knowledge in the head
• Simplify the structure of tasks
• Make things visible: bridge the gulfs of Execution and Evaluation
• Get the mappings right
• Exploit the power of constraints (natural and artificial)
• Design for Error
• When all else fails, standardize.

• An understanding of cognitive design can assist in creating high-quality, user-centered objects, interfaces and multimedia.
• Familiarise yourself with the concepts above and look around you to see how well (or how poorly) they have been used in the artefacts you operate.