Cognitive Model

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 65

Cognitive Model

Learning outcomes
 Measure the response time based on the KLM
 Identifying goals, operators, methods and selection
rules using GOMS.
GOMS
 Many models make use of a model of mental processing
in which the user achieves goals by solving subgoals in
a divide-and-conquer fashion.
 The GOMS model of Card, Moran and Newell is an
acronym for Goals, Operators, Methods and
Selection
• Goals:
– Something that the user tries to accomplish (action-object pair, e.g.
delete word)
– Include context
• Methods:
– Well learned sequence of steps that accomplish a task
– How do you do it on this system? (could be long and tedious…)
• Selection Rules:
– Only when there are clear multiple methods for the
same goal.
• Operators:
– Elementary perceptual, cognitive and motor acts
that cause change (external vs. mental)
– Also uses action-object pair (e.g. press key, select
menu, make gesture, speak command...)
– mostly defined by hardware and lower-level software.
Why Do We Use The GOMS Models?
 the ability to make a priori predictions;
 the ability to be learned and used by practitioners as
well as researchers;
 coverage of relevant tasks;
 approximation.
What can GOMS model?
• Task must be goal-directed
– Some activities are more goal-directed than
others
– Even creative activities contain goal directed tasks
• Task must be a routine cognitive skill
• Can include serial and parallel tasks
Cont…

Send a message to friend


Cont…
Cont…
Close Window
Examples

• Travel from Jaipur to Newyork


• Report a Product Issue
• Shut down a system
GOMS model – types
 Keystroke-Level Model (KLM)
 CMN-GOMS (Card Moran Newell)
 NGOMSL (Natural GOMS language)
 CPM-GOMS (Critical Path Model)
Keystroke Level Model (KLM)
 Why Use the KLM?
 Consider a task such as “delete a file”
 Perhaps there are two ways to do the task:
 Mouse + menu selection
 Keyboard + command entry
 The KLM can predict the time for each method
 If used at the design stage, design alternatives may
be considered and compared design choices follow
Advantages of Predictive Evaluation
 Don’t have to build UI prototype
 Can compare design alternatives with no implementation
whatsoever
 Don’t have to test real live users
 Theory provides explanations of UI problems
 So it points to the areas where design can be improved
 User testing may only reveal problems, not explain them
How to use KLM to measure and compare the efficiency
of user interfaces?

 Make predictions of the task execution times for


a specific design
 Efficiency is measured as the speed with which a user can
accomplish the task.
A KLM Prediction
 A task is broken into a series of subtasks
 Total predicted time is the sum of the subtask
times:
T(execute) = ∑(time to execute primitive op)
 Primitive operations
– K key/button press
– P point to target with mouse
– H home hands to keyboard or mouse
– D draw line with mouse
– M mental preparation (pause)
– R system response time
 Times for primitive operations are predicated from
experiments.
 Time to press key/button ranges between
• 0.08 fast typist
sec/char average typist
• 0.28 slow typist
sec/char
• 1.2 sec/char
KLM Analysis

 Encode a method as a sequence of physical operators


(KPHD)
 Use heuristic rules to insert mental operators (M)
 Add up times for each operator to get total time
for method
how to create a keystroke level model for a
task
 First focus on a particular method for doing the task.
Suppose the task is deleting a word in a text editor. Most
text editors offer a variety of methods for doing this, e.g.
1. click and drag to select the word, then press the Del key
2. click at the start and shift-click at the end to select the
word, then press the Del key
3. click at the start, then press the Del key N times
4. double-click the word, then select the Edit/Delete
menu command; etc.
 Second, encode the method as a sequence of the
physical operators: K for keystrokes, B for mouse
button presses or releases, P for pointing tasks, H for
moving the hand between mouse and keyboard, and D
for drawing tasks.
 Third, insert the mental preparation operators at the
appropriate places, before each chunk in the task.
Some heuristic rules have been proposed for finding
these chunk boundaries.
 finally, using estimated times for each operator, add up
all the times to get the total time to run the whole
method.
Heuristic Rules for adding M’s
 Basic idea:
 M before every chunk in the method that must be recalled from long-
term memory or that involves a decision
 Initiating a task.
 The user has to pause and make a definite decision about what the task is,
and what should be done.Thus users often pause before emitting a sequence
of actions; this pause should be routinely represented by an M
 Making a strategy decision.
 If there is more than one way to proceed, and the decision is not obvious or
well practiced, but is important, the user has to stop and think.
 Finding something on the screen.
 The user must pause and scan the screen for an item that they do not already
know or whose location on the screen they do not already know from
practice.
 Retrieving a chunk from memory
 Command name
 File name
 Parameter value
 Verifying that a specification or action is correct.
 Before users signal the system to proceed, they often pause
and check their entry. For example, users often stop and
examine a command before hitting return, or check a dialog
box before clicking on the OK, or that a destination folder is
reverse-video before releasing the mouse button
 Pointing to an object on the screen should be
preceded by a mental operator to locate the object.
 If something on the screen changes in response to user
input, there should be a step to veryify that the
desired result appeared.
 Before every keypress. And if the keys are
consecutive then one M operator is placed.
Example
 keystroke-level models for two methods that delete a
word.
Method 1:
 The first method clicks at the start of the word, shift-clicks
at the end of the word to highlight it, and then presses the
Del key on the keyboard.
Method 2:
 The second method clicks at the start of the word, then
presses Del enough times to delete all the characters in
the word.
K - 0.28 s average typist
 B - 0.1 s
 P - 1.1 s
 H – 0.4 s
 M – 1.2 s
Method 1 - Shift-click selection
 M
 P [start of word]
 BB [click]
 H
 M
 K [shift]
 M
 P [end of word]
 BB [click]
 H [to keyboard]
 M
 K [Del]
 M

Total: 5M + 2P + 4B +
2K
Method 2 - pressing Del key n times
 M
 P [start of word]
 BB [click]
H
M
 K [Del] × n [where n = length of
word]
M

Total:
3M+P+2B+H+nK
Comparing efficiency of different user
interface designs
parametric analysis
 e.g., as we vary the parameter n (the length of the word to
be deleted), how do the times for each method vary?
Consider the following three different interfaces for collecting the date of birth.
Calculate the time taken to perform each of these operations using a KLM and
perform a task execution time analysis to find which interface model is a best
alternative for collecting DOB. List the design heuristics that should be considered
for placing the M operator for the following interface designs.

K - 0.28 s average
typist
B - 0.1 s
P - 1.1 s
H – 0.4 s
M – 1.2 s
CMN GOMS
Suppose we want to find out the definition of a word
from an online dictionary. How can we model this task
with (CMN)GOMS?

 list the goals (high level tasks) first


– Goal: Access online dictionary (first, we need to access the
dictionary)
– Goal: Lookup definition (then, we have to find out the
definition)
 Next, we have to determine the methods (operator or
goal- operator sequence) to achieve each of these goals
Goal: Access online dictionary
Operator:Type URL sequence
Operator: Press Enter
Goal: Lookup definition
Operator:Type word in entry
field
Goal: Submit the word
Operator: Move cursor from field to Lookup
button
Operator: Select Lookup
Operator: Read output
 The previous example illustrates the concepts of goals
and goal hierarchy, operators and methods
 The other important concept in ( C M N ) G O M S is
the selection rules
Suppose we have a window interface that can be closed in
either of the two methods: by selecting the ‘close’ option from
the file menu or by selecting the Ctrl key and the F4 key
together. How we can model the task of “closing the window”
for this system using (CMN) GOMS?

Goal: Close window


[Select Goal: Use menu method
Operator: Move mouse to file
menu Operator: Pull down file
menu Operator: Click over close
option
Goal: Use Ctrl+F4 method
Operator: Press Ctrl and F4 keys
together]
Cognitive complexity theory (CCT)
 Cognitive complexity theory, introduced by Kieras and Polson.
 two parallel descriptions: one of the user’s goals and the
other
of the computer system.
 user’s goals is based on a GOMS-like goal hierarchy – expressed
in the form of production rules.
 system grammar, CCT uses generalized transition networks, a form
of
state transition network
 The production rules are a sequence of rules: (rule program is
written in LISP)
if condition then action
 Where, condition is a statement about the contents of
working memory. action may consist of one or more
elementary actions, which may be either changes to the
working memory, or external actions such as keystrokes
Example
Selection rule set for goal: Highlight text
If text-is word, then accomplish goal: Highlight word.
If text-is arbitrary, then accomplish goal: Highlight arbitrary
text.
Return with goal accomplished.
Delete a file using GOMS
ATM machine – Basic operators
 GOAL: GET-MONEY
 GOAL: USE-CASH-MACHINE
 INSERT-CARD
 ENTER-PIN
 SELECT-GET-CASH
 ENTER-AMOUNT
 COLLECT-CARD
 COLLECT-MONEY (outer goal satisfied!)
Natural GOMS
 Natural GOMS language (NGOMSL), developed by
Kieras, provides a structured natural-language
notation for GOMS analysis and describes the
procedures for accomplishing that analysis.
 NGOMSL provides
 A method for measuring the time it will take to learn
specific method of operation
 A way to determine the consistency of a design’s methods
of
operation
Method for goal: Cut text
Step 1. Accomplish goal: Highlight text.
Step 2. Return that the command is CUT, and accomplish goal: Issue a
command.

Step 3. Return with goal accomplished. ...

Selection rule set for goal: Highlight text

If text-is word, then accomplish goal: Highlight word.


If text-is arbitrary, then accomplish goal: Highlight arbitrary text. Return
with goal accomplished. ...

Method for goal: Highlight arbitrary text


Step 1. Determine position of beginning of text (1.20
sec) Step 2. Move cursor to beginning of text (1.10 sec)
Step 3. Click mouse button. (0.20 sec)
Step 4. Move cursor to end of text. (1.10 sec)
Step 5. Shift-click mouse button. (0.48 sec)
Step 6.Verify that correct text is highlighted (1.20
sec) Step 7. Return with goal accomplished.

This NGOMSL model predicts that it will take 5.28


seconds to highlight
arbitrary text
CPM - GOMS
 CPM – Critical Path Model
 CPM-GOMS builds on previous GOMS models by
assumed that perceptual, cognitive and motor operators
can be performed in parallel
 It employs a schedule chart (also known as a PERT chart)
to
represent operators and dependencies between
operators.
 Example, the goal READ-SCREEN, when an eye movement
is required.
 The sequence which produces the longest path through the
chart is called the critical path, and it represents an estimate
of the total time required to perform the task.
Point-Shift-clickoperation - the path that takes the
longest time, since it will determine the total time for
the method.

Dheeba. J,
SCOPE
CPM GOMS model for Move text method
Modelling structure – Hick’s law
 Hick’s Law (or the Hick-Hyman Law) is named after
a British and an American psychologist team of
William Edmund Hick and Ray Hyman
 Hick’s law (Hick-Hyman law) can help to optimize menu
structures.
 Hick’s Law states that increasing the number of
choices
will increase the decision time logarithmically.
 The reaction time has a logarithmic curve, since users
can process information presented in categories or
groups, therefore when the user makes a choice, he/she
eliminates whole groups of other choices — that is why
the curve is not linear.
 As a designer, you will use Hick’s Law to examine
how many functions you should offer at any part of
your website and how this will affect your users’
overall approach to decision making.
 The formula for Hick’s Law is defined as follows:

𝑅𝑇 = 𝑎+ 𝑏log2 (𝑛 + 1)
 The coefficients a and b are arrived at from
experiments which depend on the person, training,
other factors.
 If the number of choices is related with probability
of occurances
H = 𝑃𝑙𝑔 (𝑗 1 + 21)𝑃𝑗
𝑜
 Hick’s Law determined the number of controls on your
microwave or your washing machine.
 A design principle known as “K.I.S.S.” (“Keep It Short and
Simple”) became recognized in the 1960s for its effectiveness
in this regard.
 Categorizing Choice:
 You can see Hick’s Law in action in the navigation of almost any
website.
 If your menus offered direct access to every link within your site,
you could quickly overwhelm the visitor.
 Obscuring Complexity
 If you have a complex process, you can use Hick’s Law to rationalize
only presenting specific parts of that process at any one time on
the screen.
 Instead of throwing the entirety of your payment process up in a
long, complex form, you can break it down into prompting users
to register their e-mail and create a password.
 Then, you can give them another screen with shopping cart
details, then another which collects delivery information and so
on.
 Amazon gives you these
options to find what you are
looking for with ease.
 Though the options are
large it is broken into
groups
 Imagine the amount of
options that will be available
on amazon if they don’t
follow the hick’s law
Case study – Hick’s Law
 Zomato is one of the biggest online food ordering
platforms today. In a nutshell it is the biggest and widest
menu card a person has to choose food from. So
naturally, the choices are not only ample but each more
delicious than the other. Zomato does offer a range of
filtering options to narrow down our food choices, such
as — delivery time, cost of the food, cuisine, offers and
many more. Even with the availability of these filters,
users can take up to 30 minutes to decide what to
order, which gets more confusing if you are deciding as a
group. So how do we apply hick’s law here.
 delicious looking fresh food pictures, attached with every
restaurant and their menu cards. The user is constantly
browsing through menu items while looking at these
mouth-watering images.
 So, is it a coincidence that the user is presented with
numerous choices, thereby increasing the decision-making
time while constantly exposing them to food pictures? If
it is, Zomato has cracked the code to smartly getting
more orders from one user without having to persuade
them any other way. After all, a hungry user is a user
they can serve well.
 So, the crux here is, choices are not always harmful for a
company’s profit.
Modelling Dynamics - State Transition
Network (STN)
 Interaction designs involve dynamic feedback loops
between the user and the system
 User actions alter the state of the system, which in
turn
influences the user’s subsequent actions.
 Interaction designers needs tools to explore how a
system
undergoes transition from one state to another.
 STN helps to explore
 Menus
 Icons
 Tools
 STN helps to show the operation of the
peripheral devices.
 STNs are the most spontaneous, which knows that a
dialog fundamentally denotes to a progression from
one state of the system to the next.
 The syntax of an STN consists of the following
two entities
 Circles − A circle refers to a state of the system, which
is branded by giving a name to the state.
 Arcs − The circles are connected with arcs that refers to
the action/event resulting in the transition from the state
where the arc initiates, to the state where it ends.
 STNs are appropriate for showing sequential operations
that may involve choice on the part of the user as well
as for expressing iterations.
Structure of STN
Modelling dynamics – three state model
 The three state model can help designers to determine
appropriate I/O devices for specific interaction
designs.
 All input devices are not created equal in their
capabilities, nor input techniques (such as pointing,
dragging, and rubber-banding) in their demands.
 Variations are in terms of qualitative (types of signal) and
quantitative (amount of signal)
 Consider moving the mouse without the button
pushed.
 One way to characterize the state of the system at this
point is as tracking.
 If, however, we point at an icon, depress the mouse
button and move the mouse while holding the
button down, we have entered a new state,
dragging.

Simple 2-State Transaction


 Consider now the situation if a touch tablet rather than a
mouse was connected to the system.

 For the purpose of the example, let us assume that the


touch tablet is capable of sensing only one bit of
pressure, namely touch or no-touch.
 The first state, (State 0), is what we will call out of
range, (OOR).
 In this state, any movement of the finger has no effect
on the system. It is only when the finger comes in
contact with the tablet that we enter the State 1.

State 0-1 Transaction


 Assume a graphics tablet with stylus.
 In State 0, the stylus is off of the tablet and the tip switch
in its open state.
 Moving the stylus has no effect since it is out of range
(OOR).
When the stylus is in range, the tracking symbol follows
the stylus' motion (State 1: tracking).
 Extra pressure on the stylus closes the tip switch,
thereby moving the system into State 2.

State 0-1-2 Transaction


Modelling dynamics – Glimpse Model
 Technique that supports the previewing of navigation,
exploration, and editing operations by providing convenient
Undo for unsuccessful and/or undesirable actions on multi-
level input devices such as touch screens and pen-based
computers.
 By adding a Glimpse state to traditional three-state pressure
sensitive input devices, users are able to preview the effects of
their editing without committing to them.
 Creativity is enhanced when users are able to easily retract
their changes if the result is unsatisfactory.
 multi-level input devices - input device is capable of sensing at
least two levels of input
 e.g. a stylus that senses light and heavy pressure or a mouse with a
two-state button
 providing positional feedback
 Onscreen or implicit
 three-state model for pressure sensitive input
 light pressure input results in the “tracking” of the
input device (similar to moving the mouse while its
button is up).
 Heavy pressure input results in “dragging” operations
(similar to moving the mouse while its button is down)
Applications
Physical Models – Fitts Law
 named after Paul Fitts, for his study of
pointing.
 pointing with mouse, using touch screens and
the range of pointing devices for 2D displays.
 the law tells us how long it will take to move
a pointer from a specific position to hit
different targets.
 targets that are larger and closer are easier
to hit than ones that are smaller and farther
away.
 we can also use Fitts' law to compare different
input devices.
Fitts’ Law – Formula
The time to acquire a target is a function of the distance
to and size of the target and depends on the particular
pointing system

MT: movement time


a and b: constants dependent on the pointing
system
D: distance to the target area
W: width of the target

Dheeba. J,
SCOPE
Index of Difficulty

 MT = a + b ·ID
 ID describes the difficulty of the task
 independent of the device / method
 Index of Difficulty, ID measured in bits
 Throughput - index of performance or bandwidth
 Single metric for input systems
 One definition:TP = ID / MT (‘average’ values of ID and
MT
are used)
 Another definition:TP = 1 / b (equals ID / MT only if a=0)
Design implications – Fitts’ law
 Bigger buttons
 e.g. web links
 e.g. check / radio boxes
 any other interactive element in the graphical user interface
must be distinguished from other non-interactive elements
by size.
 Use edges and corners (for examples see next slide)
 edges of the screen have infinite height or width, respectively
 corners have infinite height and width.
 As the user is restricted in their movements the
pointing device cannot move any further when they
reach the outermost points of the screen
Pop-up Menus
 These menus support immediate selection of interactive
elements than dropdown menus as the user does not have
to move the cursor from its current position.

Muscular tension
 mobile devices are often carried around in pockets, which can
trigger commands by accident. In those situations, high-
precision input methods are deployed, which use a higher input
difficulty to make sure that a command is not executed
accidentally.
References
 Nptel on human computer interaction
 Alan Dix, Human Computer Interaction:
Practice
Thank You

You might also like