Simple Models of Human Performance

1
Simple Models of Human Performance

• Predictive Evaluation with Hicks Law, Fitts
  Law, Power Law of Practice, Keystroke-Level Model

This material has been developed by Georgia Tech
HCI faculty, and continues to evolve.
Contributors include Gregory Abowd, Jim Foley,
Diane Gromala, Elizabeth Mynatt, Jeff Pierce,
Colin Potts, Chris Shaw, John Stasko, and Bruce
Walker. Comments directed to foley_at_cc.gatech.edu
are encouraged. Permission is granted to use with
acknowledgement for non-profit purposes. Last
revision May 2004.
2
Simple User Models

• Idea If we can build a model of how a user
  works, then we can predict how s/he will interact
  with the interface
• Predictive model ? predictive evaluation
• No mock-ups or prototypes!

3
Two Types of User Modeling

• Stimulus-Response
• Hicks law
• Practice law
• Fitts law
• Cognitive human as interperter/predictor
  based on Model Human Processor (MHP)
• Key-stroke Level Model
• Low-level, simple
• GOMS (and similar) Models
• Higher-level (Goals, Operations, Methods,
  Selections)
• Not discussed here

4
Power law of practice

• Tn T1n-a
• Tn to complete the nth trial is T1 on the first
  trial times n to the power -a a is about .4,
  between .2 and .6
• Skilled behavior - Stimulus-Response and routine
  cognitive actions
• Typing speed improvement
• Learning to use mouse
• Pushing buttons in response to stimuli
• NOT learning

5
How can we use this law?

• Watching on web? Stop and reflect andmake
  notes.
• In class?Discussion time.

6
Uses for Power Law of Practice

• What did you think of?
• Use measured time T1 on trial 1 to predict
  whether time with practice will meet usability
  criteria, after a reasonable number of trials
• How many trials are reasonable?
• Predict how many practices will be needed for
  user to meet usability criteria
• Determine if usabiltiy criteria is realistic

7
Hicks law

• Decision time to choose among n equally likely
  alternatives
• T Ic log2(n1)
• Ic 150 msec

8
How can we use this law?

• Watching on web? Stop and reflect
• In class? Discussion time

9
Uses for Hicks Law

• What did you think of?
• Menu selection
• Which will be faster as way to choose from 64
  choices? Go figure
• Single menu of 64 items
• Two-level menu of 8 choices at each level
• Two-level menu of 4 and then 16 choices
• Two-level menu of 16 and then 4 choices
• Three-level menu of 4 choices at each level
• Binary menu with 6 levels

10
Fitts Law

• Models movement times for selection (reaching)
  tasks in one dimension
• Basic idea Movement time for a selection task
• Increases as distance to target increases
• Decreases as size of target increases

11
Original Experiment

• 1-D

d
w
12
Components

• ID - Index of difficulty
• ID is an information theoretic quantity
• Based on work of Shannon larger target more
  information (less uncertainty)

ID log2 (d/w 1.0)
width (tolerance) of target
bits result
distance to move
13
Components

• MT - Movement time
• MT is a linear function of ID
• k1 and k2 are experimental constants

MT k1 k2ID MT k1 k2 log2 (d/w 1.0)
14
Exact Equation

• Run empirical tests to determine k1 and k2 in MT
  k1 k2 ID
• Will get different ones for different input
  devices and device uses

MT
ID log2(d/w 1.0)
15
Questions

• What do you do in 2D?
• h x l rectone way is ID log2(d/min(w, l) 1)
• Should take into account direction of approach

16
How can we use this law?

• Watching on web? Stop and reflect andmake
  notes.
• In class?Discussion time.

17
Uses for Fitts Law

• What did you think of?
• Menu item size
• Icon size
• Scroll bar target size and placement
• Up / down scroll arrows together or at top and
  bottom of scroll bar

18
Keystroke-Level Model (KSLM)

• KSLM - developed by Card, Moran Newell, see
  their book and CACM
• The Psychology of Human-Computer Interaction,
  Card, Moran and Newell, Erlbaum, 1983
• Skilled users performing routine tasks
• Assigns times to basic human operations -
  experimentally verified
• Based on MHP - Model Human Processor

19
KSLM Accounts for

• Keystroking TK
• Mouse button press TB
• Pointing (typically with mouse) TP
• Hand movement betweenkeyboard and mouse TH
• Drawing straight line segments TD
• Mental preparation TM
• System Response time TR

20
Using KSLM - Step One

• Decompose task into sequence of operations - K,
  B, P, H, D (no M operators yet R can be used
  always or not at all)

21
Step One MS Word Find Command

• Use Find Command to locate a six character word
• H (Home on mouse)
• P (Edit)
• B (click on mouse button - press/release)
• P (Find)
• B (click on mouse button)
• H (Home on keyboard)
• 6K (Type six characters into Find dialogue box)
• K (Return key on dialogue box starts the find)

22
Using KSLM - Step Two

• Place M operators
• Rule 0a. In front of all Ks that are NOT part of
  argument strings (ie, not part of text or
  numbers)
• Rule 0b. In front of all Ps that select commands
  (not arguments)

23
Step Two MS Word Find Command

• H (Home on mouse)
• MP (Edit)
• B (click on mouse button)
• MP (Find)
• B (click on mouse button)
• H (Home on keyboard)
• 6K (Type six characters)
• MK (Return key on dialogue box starts the find)

Rule 0b Pselects command
Rule 0b Pselects command
Rule 0a Kis argument
24
Using KSLM - Step 3

• Remove Ms according to heuristic rules
• (Rules relate to chunking of
  actions)
• Rule 1. Anticipated by prior operation
• PMK -PK (point and then click is a chunk)
• Rule 2. If string of MKs is a single cognitive
  unit (such as a command name), delete all but
  first
• MKMKMK - MKKK (same as M3K) (type run rtn is a
  chunk)
• Rule 3. Redundant terminator, such as )) or rtn
  rtn
• Rule 4. If K terminates a constant string, such
  as command-rtn, then delete M
• M2K(ls)MK(rtn) - M2K(ls)K(rtn) (typing ls
  command in Unix followed by rtn is a chunk)

25
Step 3 MS Word Find Command
H (Home on mouse) MP (Edit) B (click on mouse
button) MP (Find) B (click on mouse button) H
(Home on keyboard) 6K (Type six characters) MK
(Return key on dialogue box starts the find)
Rule 1 delete M H anticipates P
Rule 1 delete M H anticipates P
Rule 4 Keep M
26
Using KSLM - Step 4

• Plug in real numbers from experiments
• K .08 sec for best typists, .28 average, 1.2 if
  unfamiliar with keyboard
• B down or up - 0.1 secs click - 0.2 secs
• P 1.1 secs
• H 0.4 secs
• M 1.35 secs
• R depends on system often less than .05 secs

27
Step 4 MS Word Find Command

• H (Home on mouse)
• P (Edit)
• B (click on mouse button - press/release)
• P (Find)
• B (click on mouse button)
• H (Home on keyboard)
• 6K (Type six characters into Find dialogue box)
• MK (Return key on dialogue box starts the find)
• Timings
• H 0.40, P 1.10, B 0.20, M 1.35, K 0.28
• 2H, 2P, 2B, 1M, 6K
• Predicted time 6.43 secs

28
Example MS Windows Menu Selection

• Get hands on mouse
• Select from menu bar with click of mouse button
• The pull down menu appears
• Select desired item from the pull down menu

29
Step 1 MS Windows Menu

• H (Home on mouse)
• P (point to menu bar item)
• B (left-click with mouse button)
• P (point to menu item)
• B (left-click with mouse button)

30
Step 2 MS Windows Menu - Add Ms

• H (get hand on mouse)
• MP (point to menu bar item)
• B (left-click with mouse button)
• MP (point to menu item)
• B (left-click with mouse button)

Rule 0b Pselects command
Rule 0b Pselects command
31
Step 3 MS Windows Menu - Delete Ms

• H (get hand on mouse)
• MP (point to menu bar item)
• B (left-click with mouse button)
• MP (point to menu item)
• B (left-click with mouse button)

Rule 1 Manticipated by P
Keep M
32
Step 4 MS Windows Menu Calculate Time

• H (get hand on mouse)
• P (point to menu bar item)
• B (left-click with mouse button)
• MP (point to menu item)
• B (left-click with mouse button)
• Textbook timings (all in seconds)
• H 0.40, P 1.10, B 0.20, M 1.35
• H, 2P, 2B, 1 M
• Total predicted time 4.35 sec

33
Macintosh Menu Selection

• Operator sequence
• H(mouse)P(to menu item)B(down)PB(up)
• Now place Ms
• H(mouse)MP(to menu item)B(down)MPB(up)
• Selectively remove Ms
• H(mouse)MP(to menu item)B(down)MPB(up)
• Textbook timings (all in seconds)
• H 0.40, P 1.10, B 0.10 for up or down, M
  1.35
• H, 2P, 2 B, 1 M
• Total predicted time 4.15 sec
• Macintosh is predicted to be .2 secs faster than
  MS Windows, about 5

Rule 0b
Rule 0b
Rule 1 Delete H anticipates P
34
KSLM Comparison Problem

• Are keyboard accelerators always faster than menu
  selection?
• Use MS Windows to compare
• Menu selection of File/Print (previous example
  estimated 4.35 secs.)
• Keyboard accelerator
• ALT-F to open the File pull down menu
• P key to select the Print menu item
• Assume hands start on keyboard

35
KSLM ComparisonKeyboard Accelerator for Print

• Use Keyboard for ALT-F P (hands already there)
• K(ALT)K(F)K(P)
• MK(ALT)MK(F)MK(P)
• MK(ALT)K(F)MK(P)
• 2M 3K 2.7 3K
• Times for K based on typing speed
• Good typist, K 0.12 s, total time 3.06 s
• Poor typist, K 0.28 s, total time 3.54 s
• Non-typist, K 1.20 s, total time 6.30 s
• Time with mouse was 4.35 sec
• Conclusion Accelerator keys not necessarily
  faster than mouse!

First K anticipates second K
36
KSLM Example - select a word and replace with new
typed text

• Home on mouse H(mouse)
• Point to word P(word)
• Select word BB(mouse button)
• Home on keyboard H
• Type new word KKKKK

37
KSLM Example

• No Ms to add
• Ks are part of argument, so rule 0a does not
  apply
• No Ps to use with rule 0b
• Sequence remains as
• Home on mouse H(mouse)
• Point to word P(word)
• Select word BB(mouse button)
• Home on keyboard H
• Type new 5-letter word 5K
• T 5TK 2TB TP 2TH TM 5(.28)2(.2)1.12(.4)
  1.35 5.05 secs

38
Using KSLM

• Skilled users
• Performing routine tasks
• The user has done it many times before
• No real learning going on
• Some modest thinking as captured by Ms
• Rules for placing Ms are heuristics
• Best use is for comparing alternatives
• Sometimes predictions are off
• But rankings of faster - slower tend to be
  accurate

39
Now You Get to Do It

• Draw through text and make it bold
• By pointing to BOLD icon in floating palette
• By selecting BOLD from pull-down menu
• Revisit the hierarchical menu selection example,
  do a full KSLM analysis
• Now use Hicks Law at appropriate places

40
Cognitive models - many flavors

• More complex than KSLM
• Hierarchical
• GOMS - Goals, Operators, Methods, Selectors
• CCT - Cognitive Complexity Theory
• Linguistic
• TAG - Task Action Grammar
• CLG - Command Language Grammar
• Cognitive architectures
• SOAR, ACT

41
End