Schneiderman Interaction styles Human error Norman - PowerPoint PPT Presentation

1 / 34
About This Presentation
Title:

Schneiderman Interaction styles Human error Norman

Description:

Music. Natural. Synthetic or sampled. The ears tell the eyes where to look ... Pop up menus. Appear at various places on the screen ... – PowerPoint PPT presentation

Number of Views:44
Avg rating:3.0/5.0
Slides: 35
Provided by: elizabet61
Category:

less

Transcript and Presenter's Notes

Title: Schneiderman Interaction styles Human error Norman


1
SchneidermanInteraction stylesHuman error
(Norman)

2
Direct manipulation design principles
  • Visually represent the world of action
  • Objects of interest shown on screen
  • Actions of interest shown on screen
  • Rapid execution of actions
  • Actions s/b reversible
  • Interaction accomplished through
  • Pointing
  • Selecting
  • Dragging

3
Direct manipulationWhy does it work?
  • According to Schneiderman
  • It engages human perceptual recognition
  • Human vision is an effective way to learn and
    understand the world

4
Visualization Tree maps
  • A way to visually explore data
  • In this case it is a visual representation of
    3200 files
  • Rectangle size reflects the file size
  • Works well if your goal is to identify which file
    (or files) are the biggest disk hogs
  • Pointing at a rectangle uncovers more information
    about that particular file
  • Question What do the colors represent?

5
Skipped Chapters 11 12
  • Interesting stuff but most of you guys know about
    this stuff
  • I will briefly cover Sound Output starting on
    page 247...

6
Sound Output
  • Kinds of sounds
  • Speech
  • Music
  • Natural
  • Synthetic or sampled
  • The ears tell the eyes where to look
  • Excellent for monitoring the status of some
    process not being directly attended to

7
Natural soundsAudio Icons
  • Human hearing has evolved to allow the extraction
    of a great deal of information from the nature of
    the sounds heard
  • The sounds generated tell us about the properties
    of those things making the sounds
  • But then there is the Foley artist recording a
    sound that generated by the real thing isnt
    necessarily sufficient. Example gunshot sounds
    like a cracked whip when recorded...

8
Natural soundsExample of use with email
  • Scott Hudson paper on doing exactly this
  • http//www.acm.org/sigchi/chi96/proceedings/shortp
    ap/Hudson/hs_txt.htm

9
Musical sounds
  • Earcons
  • Distinctive musical sounds that are associated
    with particular operations
  • Examples
  • http//www.dcs.gla.ac.uk/stephen/earconexperiment
    1/earcon_expts_1.shtml

10
Speech
  • Speech synthesis technology
  • Concatenation (Less flexible vocabulary)
  • Uses recorded sounds that are concatenated
  • Synthesis by rule (More flexible)
  • Rules control the concatenation of phonemes
  • Phonemes can be recorded
  • Phonemes can be generated by a model of the human
    vocal tract (Show Dude)

11
Chapter 13 Interaction styles
  • Interaction styles include
  • Turn-taking dialogs
  • Take the general form
  • User specifies instructions
  • System executes the instructions
  • System provides feedback
  • Command line entry
  • Manipulation of objects and tools
  • Gesturing

12
Interaction styles
  • Command line entry (nuf said)
  • Form filling and spreadsheet (nuf said)
  • Menus and navigation (more to come--gt )
  • Natural language (more to come--gt )
  • Direct manipulation (more to come--gt )

13
Menus
  • Types
  • Pull down, pop up, pie menu
  • Take advantage of recognition rather than memory
  • Menu organization
  • Alphabetical
  • Chronological
  • Categorical

14
Menus (cont)
  • Provide a major real-estate savings
  • Pull down menus
  • Location is permanently set ()
  • Location is permanently set (-)
  • Pop up menus
  • Appear at various places on the screen
  • User needs to be given an indication of their
    availability
  • Pie menus
  • Typically appear around the cursor
  • Can obscure some of the screen

15
Natural language dialogue
  • Spoken vs keyboard entry
  • Reduces ambiguity by removing recognition from
    the picture
  • Ambiguity is a real problem
  • Limits implementation to some subset of natural
    language
  • Subsets have been used to build expert systems
    and intelligent tutors

16
Direct manipulation
  • (lt-- See Schneiderman discussion earlier)
  • Novices gain basic functionality quickly
  • Users feel in control of the system
  • Users experience less anxiety because the system
    is comprehensible and reversible
  • Users see the results of their actions and can
    determine if they are furthering their goals

17
The gulfs (Norman)
  • The gulf of execution
  • The distance between the users goals and the
    means of achieving them using the system
  • The gulf of evaluation
  • The distance between what the system presents as
    feedback and the users expectations of that
    feedback

18
Bridging the gulfs
  • The user can change
  • Adapt to the system (faulty design?)
  • The designers can change the system
  • To better match the expectations of the user
  • Create a means of input that better matches the
    users expectations of how to use the system
  • Create displays of system feedback to better
    match the users expectations of that feedback

19
The notion of directness expanded Semantic
directness
  • Does the interface support what the user wants to
    say in a concise way?
  • If the user wants to send a document that
    includes text, pictures, audio and video can he
    do it directly or must he create 4 different
    documents?

20
The notion of directness expanded Articulatory
directness
  • Similar to the idea of mapping
  • Example The turn signal that pushes forward and
    pulls back, when mapped to the idea of pushing
    and pulling on the handlebar, gets a high rating
    in articulatory directness

21
Affordances and constraints
  • Affordances of an object suggest the range of
    possibilities for use of that object
  • Constraints limit the range of possibilities
  • Physical constraints shape is an example
  • Semantic constraints meaning constrains choice
  • Cultural constraints ingrained, well-learned
    meanings, schemata
  • Logical constraints only one place left

22
DOET 5 Human error
  • Mistakes and slips (recap)
  • reference the notes from Preece, chapter 8
  • Mistakes are the result of conscious deliberation
  • Slips result from automatic behavior
  • Capture, description, data-driven, associative
    activation, loss-of-activation and mode errors

23
More on Slips
  • Easy to detect discrepancy between intended
    result and the actual result
  • Only if results of the action are visible
  • Actions are specified at different levels at
    which level did the error occur?

24
How can you handle slips?
  • Prevent slips from happening by designing
    appropriately
  • For example, if you have modes, reduce the number
    or eliminate them by redesign
  • Make things deliberately different
  • Detect and correct them when they do
  • Confirmation only handles part of the problem,
    with deletion of a file the user is confirming
    the action (delete) not the filename (which is in
    error)
  • Provide for recovery
  • Trash can does not delete, rather stores for
    deletion

25
One small problem
  • When you design an error-tolerant system, people
    come to rely on that system! (It had better be
    RELIABLE!)
  • Anti-skid brakes
  • Blade guard on circular saw

26
Human thought
  • Is dependent on the underlying memory
  • Is not like logic, it is much messier
  • Memory
  • Not like a photo album, not that neat and orderly
  • Filing cabinet models of memory
  • Schema, frames, semantic networks, propositional
    encoding,
  • Connectionist
  • Represented as patterns of activation and
    inhibition
  • Much is hidden, inaccessible, beneath the
    surface, with only the end states available for
    conscious inspection

27
Task structure
  • Descriptions of even simple tasks result in wide
    AND deep structures
  • Consider the game of tic-tac-toe
  • Shallow
  • Many, simple choices, few steps
  • Narrow
  • Few choices, many steps

28
Everyday tasks
  • Not frequently studied by psychologists
  • Generally routine, requiring little conscious
    thought or planning
  • Much of human behavior is subconscious
  • Conscious thought
  • Is slow and serial
  • Is limited by the small capacity of STM
  • The nature of explanations
  • Easier to predict result AFTER it happens!
  • Obvious only happens after the fact

29
To increase errors, add a little
  • Social pressure
  • Time pressure
  • Economic pressures

30
Designing for error
  • Design to minimize error by understanding the
    causes of errors
  • Make actions reversible
  • Make error discovery easier
  • Make error correction easier
  • Change the attitude toward error from stupid
    user to stupid design

31
Dealing with error
  • Warning signals should not occur frequently to
    attract attention (key in ignition buzzer)
  • Warning signals dont work when there is the
    possibility that there will be many at the same
    time

32
Forcing functions
  • Physical constraint
  • Three flavors
  • Interlocks force a particular sequence
  • Lockins prevent premature stops (word editor)
  • Lockouts prevents entrance (stairs to basement)

33
Resultant design philosophy
  • Put knowledge in the world
  • Use natural and artificial constraints
  • Make stuff VISIBLE!
  • Narrow the gulf of execution
  • Make options visible
  • Narrow the gulf of evaluation
  • Make the results of actions visible

34
The End!
Write a Comment
User Comments (0)
About PowerShow.com