Saturday, December 11, 2010

Human Computer Interaction

Slide Set #1 Motivation and Fundamental Principles

The 7 steps of the usability engineering lifecycle

1. User working group

2. Understand user’s needs

3. Usability goals and requirements

4. Design document

5. Evaluation

6. Partner with developer to modify design

7. Iterate

The 5 perceptual depth cues

1. Size: larger objects appear to be closer

2. Interposition: blocked object appears to be behind and beyond the blocking object

3. Contrast, clarity, and brightness: sharper and more distinct objects appear to be nearer

4. Shadow: shadows cast by an object provide some cues to the relative position of objects

5. Texture: as the apparent distance increases, the texture of a detailed surface becomes less grainy

The 5 gestald laws of perceptual organization

A. Proximity

B. Similarity

C. Closure

D. Continuity

E. Symmetry

The definitions of:

1. Color stereoscopy – choosing some colors can make the unattended illusion of 3-d.

2. Usability – a term that measures the factors of interacting with a system.

3. Norman’s 3 fundamental principles

a. Controls need to be visible

i. (to the intended user, not just the designer!)

b. …with good mappings to their effects, and

c. Design should suggest (i.e., “afford”) the underlying functionality

4. Shneidermans’ principle of direct manipulation

a. Highly visible controls/results

b. Rapid, incremental, reversible, actions that remove a layer of abstraction between command and result

c. Continuous feedback

Slide Set #2 Model Human Processor + HCI Design Principles

The concepts behind the Model Human Processor, especially:

1. Recognize-Act Cycle of the Cognitive Processor

a. Analogous to fetch-execute cycle of computer processing

b. During each cycle, the contents of Working Memory initiate actions linked to related material in Long-Term Memory

c. Actions in turn modify what is in Working Memory

d. Cycle time is approximately 100 msec.

2. Variable Cognitive Processing rate

a. The cognitive processing time decreases with increased task demands and with practice

3. Discrimination

a. The harder it is to discriminate among pieces of information stored in memory, the more difficult to retrieve information

4. The difference between “pure capacity” and “effective capacity” of working memory

a. “Pure Capacity” of working memory is about 3 chunks and “Effective Capacity” uses 7 chunks and is augmented by use of Long-Term Memory

5. Power Law of Practice

a. The time Tn to perform a task on the nth trial follows a power law: Tn = T1n^-a, where a = .4

6. Uncertainty

a. Decision time T increases with uncertainty about the judgment or decision to be made

b. T = 150 * (log2(n+1)), for n equally probably alternatives: T is in milliseconds.

c. Example is what button to push in a telephone switchboard

Fitts’ Law

1. What it means – there is an equation to map out people’s ability to quickly and accurately move a device.

2. Textbook version

a. The time Tpos to move the hand to a target of size W which lies a distance A away is given by: Tpos = Im log2( A/W + .5 ) where Im = 100 msec/bit

3. Shneidermans’ Refinement

a. Assumes

i. There will always be a minimum selection time

ii. Time will be a function of the input device being used

b. Selection Time = C1 + C2(Id) + C3 ( log2(C4/W) )

i. C1 = device-dependent time to initiate a selection

ii. C2 = device-dependent time to travel to a target

iii. Id = Index of difficulty, log2(W/2 + .5 sec)

iv. W = width of the target

v. C3 = device-dependent constant

vi. C4 = device-dependent constant

Norman’s 7 stages of actions and “gulfs”

1. Actions:

a. Establish a goal

b. Form an intention

c. Specify an action sequence

d. Execute an action

e. Perceive the system state

f. Interpret the state

g. Evaluate the state with respect to their goals and intentions

2. Gulfs – these gulfs describe the gaps between the user and the interface

a. Gulf of execution

i. The difference between the user’s intentions and what the system allows them to do

b. Gulf of evaluation

i. The difference between the system’s representations and how well they can be interpreted in terms of the user’s intentions and expectations

HCI Design Principles

1. Look at interfaces and be able to cite which principles are violated, and why

a. Aesthetically pleasing

i. Provide visual appeal by following these presentation and graphic design principles

1. Provide meaningful contrast between screen elements

2. Create groupings

3. Align screen elements and groups

4. Provide 3-dimensional representation

5. Use color and graphics effectively and simply

b. Clarity

i. The interface should be visually, conceptually, and linguistically clear, including:

1. Visual elements

2. Functions

3. Metaphors

4. Words and text

c. Compatibility

i. Provide compatibility with the following:

1. The user

2. The taks and job

3. The product

ii. Adopt the user’s perspective

d. Comprehensibility

i. A system should be easily learned and understood. A user should know the following:

1. What to look at

2. What to do

3. When to do it

4. Where to do it

5. Why to do it

6. How to do it

ii. The flow of actions, responses, visual presentations, and information should be in a sensible order that is easy to recollect and place in context

e. Configurability

i. Permit easy personalization, configuration, and reconfiguration of settings to

1. Enhance a sense of control

2. Encourage an active role in understanding

f. Consistency

i. A system should look, act, and operate the same throughout. Similar components should:

1. Have a similar look

2. Have similar uses

3. Operate similarly

ii. The same action should always yield the same result

iii. The function of elements should not change

iv. The position of standard elements should not change

g. Control

i. The user must control the interaction

1. Actions should result from explicit user requests

2. Actions should be performed quickly

3. Actions should be capable of interruption or termination

4. The user should never be interrupted for errors

ii. The context maintained must be from the user’s perspective

iii. The means to achieve goals should be flexible and compatible with the user’s skills, experiences, habits, and preferences

iv. Avoid modes; they constrain actions available to users

v. Permit the user to customize aspects of the interface, while always providing a proper set of defaults

h. Directness

i. Provide direct ways to accomplish tasks

1. Available alternatives should be visible

2. The effect of actions on objects should be visible

i. Efficiency

i. Minimize eye and hand movements, and other control actions

1. Transitions between various system controls should flow easily and freely

2. Navigation paths should be as short as possible

3. Eye movement through a screen should be obvious and sequential

ii. Anticipate the user’s wants and needs whenever possible

j. Familiarity

i. Employ familiar concepts and use a language that is familiar to the user

ii. Keep the interface natural, mimicking the user’s behavior patterns

iii. Use real-world metaphors

k. Flexibility

i. A system must be sensitive to the differing needs of its users, enabling a level and type of performance based upon:

1. Each user’s knowledge and skills

2. Each user’s experience

3. Each user’s personal preference

4. Each user’s habits

5. The conditions at that moment

l. Forgiveness

i. Tolerate and forgive common and unavoidable human errors

ii. Prevent errors from occurring whenever possible

iii. Protect against possible catastrophic errors

iv. When an error does occur, provide constructive messages

m. Predictability

i. The user should be able to anticipate the natural progression of each task

1. Provide distinct and recognizable screen elements

2. Provide cues to the result of an action to be performed

ii. All expectations should be fulfilled uniformly and completely

n. Recovery

i. A system should permit:

1. Commands or actions to be abolished or reversed

2. Immediate return to a certain point if difficulties arise

ii. Ensure that users never lose their work as a result of:

1. An error on their part

2. Hardware, software, or communications problems

o. Responsiveness

i. The system must rapidly respond to the user’s requests

ii. Provide immediate acknowledgement for all user actions

1. Visual

2. Textual

3. Auditory

p. Simplicity

i. Provide as simple an interface as possible

ii. Five ways to provide simplicity

1. Use progressive disclosure, hiding things until they are needed

2. Provide defaults

3. Minimize screen alignment points

4. Make common actions simple at the expense of uncommon actions being made harder

5. Provide uniformity and consistency

q. Transparency

i. Permit the user to focus on the task or job, without concern for the mechanics of the interface

1. Workings and reminders of workings inside the computer should be invisible to the user

Chapter 9 topics of:

A. User-centered design

B. Lifecycle models

Slide Set #3 Collaboration and Localization


· Awareness and awareness support

o Awareness of other participants

§ With two participants collaborating via a computer-based application, awareness is the understanding that one has of the presence, identity and activities of the other

§ It’s an N^2 problem

o Support for awareness

§ A synchronous collaborative system should provide all participants with information about the presence, identity, and activities of all other participants – when necessary, and without anyone having to request it nor anyone specifically having to transmit it!

· The Space/Time taxonomy

o Same Place, Same Time: Synchronous, Collocated

o Same Place, Different Time: Asynchronous, Collocated

o Different Place, Same Time: Synchronous, non-Collocated

o Different Place, Different Time: Asynchronous, non-Collocated

· Reasons why collaboration applications can fail

o Users did not perceive a need to collaborate

o When the work people need to put into the application exceeds the perceived value of their benefits from using the application

o Users did not have a well-articulated collaboration “concept of operations,” including guidance for what types of communications would require what media

o Users were not available to use the application frequently enough

o Application wasn’t easy to learn/use

· Thompson’s interdependencies

o Pooled

§ Each member provides a discrete element to the whole, which consists of the pooled contributions from all

o Sequential

§ The of one person is dependent on the output of another

o Reciprocal

§ People need to resolve contingencies with others

· Differences between the 4 levels of the EWG collaboration framework

o EWG Requirement Level

§ Contents

· Work tasks

o The work that the collaborators need to do

· Transition tasks

o Tasks used to move between work tasks

· Social protocols

o Define the way in which collaborative sessions are conducted

· Group characteristics

o Attributes that determine how a group can work together

§ Metrics

· Task Outcome

· Efficiency, Scalability

· User Satisfaction

· Security

o EWG Capability Level

§ Contents

· Shared workspace

· Shared view

· N-way communication

§ Metrics

· Distribution of Participation

· Floor Control/Access

· Object Sharing Cost

· Time on Task

o EWG Service Level

§ Contents

· Whiteboards

· Application Sharing

· Text Chat

· 3D visualization

§ Metrics

· Benchmarks

· Bandwidth Throughput

o EWG Technology Level

§ Contents

· Specific Implementations

· User Interface

§ Metrics

· Quality of Service

· Audio & Video Quality

· Image Quality

· Usability Measures

· Group Characteristics that affect collaboration

o Group physical/temporal considerations

§ Are they collocated or dispersed or a mixture?

§ Are they sitting in front of a computer or walking around?

§ Are they working at the same time or different times?

§ Are some team members in other time zones?

o Group structure considerations

§ Is the group newly formed or well-established?

§ Is the group small or large, and what percentage need to collaborate at any given time?

§ Are they strongly hierarchical or egalitarian?

§ Do they have similar or dissimilar job functions?

§ Is the group homogeneous or heterogeneous (e.g., culture, computer experience, education)?

· Examples of limitations

o Lack richness of Face-to-Face

§ No non-verbal or ‘backchannel’ communication cues

§ No contextual information

o Difficult to use learned social practices to smooth interaction

§ E.g., using pauses and eye contact for effective turn-taking

§ E.g., using nods to infer conversational understanding

o When social practices are unavailable human interaction becomes challenging


· The difference between localization and internationalization

o Globalization or Internationalization

§ A basic design of “least common denominator” functionality that can act as a starting point for tailoring to specific localities

o Localization

§ Process of turning a globalized product into a product that is suited for a particular locale

§ Technical, national, and cultural aspects to localization

· Language characteristics that vary

o Script type

§ Example: Roman, Kanji, Bengali, Thai, etc…

o Number of script characters

§ Example: 26 in English vs. 22,000 in Chinese

o Multiple writing systems representing the same language

§ Example: Katakana, Hiragana, and Kanji for Japanese

o Reading direction(s) for script

§ Can be bidirectional

§ Affects insertion point

o Hyphenation

§ Note that spelling can change when word is broken!

· The fact that numbers, calendars, addresses, weights, speeds, and personal addresses may be expressed differently

Slide Set #4 Introduction to Perception and Information Visualization for HCI

The difference between models, theories, and frameworks

· Theory

o A well-substantiated explanation of some aspect of a phenomenon

o Intended to identify factors relevant to design

· Model

o A simplification of some aspect of human-computer interaction

o Intended to make it easier for designers to predict performance and evaluate alternative designs

· Framework

o A set of interrelated concepts and/or a set of specific questions

o Intended to inform or describe a particular domain area

What to consider when developing user profiles and characterizing groups

· Demographics

· Culture

· Geography

· Language

· Education

· Motivation

· Physiology

The different types of interview questions

· Open-ended

o How do you like this cell phone?

· Closed-ended

o Have you ever seen this model of cell phone before? Yes/No/Not sure

· Likert-scale

o This cell phone is just the right size for me? Agree 1 2 3 4 5 6 7 Disagree

· Semantic differential

o The size of this cell phone is: “much too small |_|_|_|_|_|_|_| much too big”

How to approach observations

· Observer unobserved by user

· Observer present with user but silent

· Observer present with user, mostly silent but asking questions on a non-interference basis

· Observer as a participant who takes part in tasks alongside operator

· Look at Environment, Work Flow, Team, Exceptional Conditions, look at Team as a whole and each member individually

What “think aloud” is all about

· Request users to verbalize thoughts while they carry out a task

o Sometimes called “think aloud” or “talk aloud”

o Normally done individually

· Be alert for information about unstated psychological processes which underlie performance, including the user’s intentions

o Will also reveal information about simple control movements

· Can use verbal protocols during observation in context or usability testing

The main concepts behind three observation and data analysis methods

· Grounded theory – another way to analyze observation data

o Develop a story that fits the data, meaning:

§ “a set of well-developed concepts related through statements of relationship, which together constitute an integrated framework that can be used to explain or predict phenomena

o Iterative process:

§ Collect data and categorize it

§ Collect more data to see if categories still make sense

§ Refine categories when new data warrants making changes

§ Quit when analysts aren’t learning anything new

o Three phases of categorizing data

§ Open coding

· Process by which categories are “discovered” in the data

· Decide on the granularity of the analysis

§ Axial coding

· Process of systematically fleshing out categories

· Relate categories to sub-categories

§ Selective coding

· Process of refining and integrating categories

· Organize categories around one central “backbone”

· Contextual inquiry – another way of gathering/analyzing data

o What: A structured, customer centered design process of collecting, interpreting, and using data for product design

o Why: Supports understanding of work, practices, interpersonal interactions, bottlenecks in workflows, use cases, etc…

o CI: Contextual Interviews, Interpretation Sessions, Affinity Diagramming, Modeling

§ Contextual Interviews

· What: One-on-one observational interviews in the workplace

· Why: Understand users through a shared interpretation of their needs, how they work, plus any issues or impediments

§ Interpretation Sessions

· What: After each interview, record each thought expressed in interview as its own note; team members make observations, ask questions, and share insights

· Why: Captures relevant data in useful form, shares interviews with cross-functional team

§ Affinity Diagramming

· What: Individual interview notes, key points, insights, questions, and design ideas grouped under themes and subthemes

· Why: Provides a way of organizing and structuring large amounts of detailed data from the bottom up

§ Modeling

· What: Different ways to organize the data obtained in the contextual interviews

· Why: Provides different windows into the data to allow analysts to view it from different angles

· Distributed cognition

o DC Overview

§ Studies the nature of cognitive phenomena across individuals, artifacts, and representations

§ Not within a person’s head, across people and what is in the environment

§ Focuses on changes in representational state

· …via transformation of info representations

§ Example: airline cockpit

· Human-human interactions

· Human-machine interactions

· Human-environment interactions

o DC Analysis

§ Show representations of orders, communications, and actions changing over time

Slide Set #5 Understanding Users

The difference between information visualization and scientific visualization

· Information visualization is the interdisciplinary study of "the visual representation of large-scale collections of non-numerical information, such as files and lines of code in software systems, library and bibliographic databases, networks of relations on the internet, and so forth"

· Scientific visualization is an interdisciplinary branch of science primarily concerned with the visualization of 3-dimensional phenomena (architectural, meteorological, medical, biological, etc.), where the emphasis is on realistic renderings of volumes, surfaces, illumination sources, and so forth, perhaps with a dynamic (time) component"

Data/attribute characteristics

  • A categorical variable, also called a nominal variable, is for mutual exclusive, but not ordered, categories. For example, your study might compare five different genotypes. You can code the five genotypes with numbers if you want, but the order is arbitrary and any calculations (for example, computing an average) would be meaningless.
  • An ordinal variable, is one where the order matters but not the difference between values. For example, you might ask patients to express the amount of pain they are feeling on a scale of 1 to 10. A score of 7 means more pain that a score of 5, and that is more than a score of 3. But the difference between the 7 and the 5 may not be the same as that between 5 and 3. The values simply express an order. Another example would be movie ratings, from * to *****.
  • An interval variable is a measurement where the difference between two values is meaningful. The difference between a temperature of 100 degrees and 90 degrees is the same difference as between 90 degrees and 80 degrees.
  • A ratio/affine variable, has all the properties of an interval variable, and also has a clear definition of 0.0. When the variable equals 0.0, there is none of that variable. Variables like height, weight, enzyme activity are ratio variables.

What “change blindness” is:

· In visual perception, change blindness is the phenomenon that occurs when a person viewing a visual scene apparently fails to detect large changes in the scene. For change blindness to occur, the change in the scene typically has to coincide with some visual disruption such as a saccade (eye movement) or a brief obscuration of the observed scene or image.

What trichromacy theory tells us about color perception (Where does the yellow come from?)

· Trichromacy is the condition of possessing three independent channels for conveying color information, derived from the three different cone types Organisms with trichromacy are called trichromats.

o The normal explanation of trichromacy is that the organism's retina contains three types of color receptors (called cone cells in vertebrates) with different absorption spectra. In actuality the number of such receptor types may be greater than three, since different types may be active at different light intensities. In vertebrates with three types of cone cells, at low light intensities the rod cells may contribute to color vision, giving a small region of tetrachromacy in the color space.

o It is estimated that each of the three cone types in the human retina can pick up about 100 different gradations, and that the brain can combine those variations such that the average human can distinguish about one million different colors.

Slide Set #6 Task Analysis

How to write a use case

· Begin with a particular user goal

· State preconditions

· Include a sequence of interactions between the actor and the system (called “narrative body”)

o Many narrative styles

o Choose the style most understandable to the anticipated users

· End when goal is satisfied

· State post conditions

How to use interview and observation data to create a hierarchical task analysis

· Start by identifying basic user tasks from observation/interview materials

· Take each high-level activity goal and…

o Identify the major steps needed to do the activity

o In cases where there are alternative ways to do the task, identify plans (sequences of steps)

o Keep decomposing steps until you get to the lowest level description that makes sense for your analysts

· Refine by working directly with users

Cognitive Task Analysis (CTA)

· Represent thought processes needed for typical tasks

· Identify info the system can provide to facilitate decision-making … and how it may need to be used

· Many CTA techniques based on hierarchical task analysis

The purpose of ACTA (Applied Cognitive Task Analysis)

· “Streamlined” form of CTA

· Based on semi-structured interviews with three sets of questions

· Researchers found that these questions provided a shortcut to learning about cognitive challenges

· After interviews with 6-12 people, analysts develop 4 analysis products

ACTA’s 4 steps

· Step 1: Task Diagram interview

o Pick a major task and ask, “can you break this task down into less than 6, but more than 3 steps?”

o Ask which steps require cognitive skill, “judgments, assessments, problem solving – thinking skills”

o Provides a surface level look at the task

· Step 2: Knowledge Audit Interview

o Identifies ways in which expertise is used in a domain and elicits actual experiences

o Example questions:

§ Is there a time when you walked into the middle of a situation and knew exactly how things got there and where they were headed?

§ Can you give me an example of what is important about the Big Picture for this task? What are the major elements you have to know and keep track of?

§ When you do this task, are there ways of working smart or accomplishing more with less – that you have found especially useful?

§ Can you think of a time when you realized that you would need to change the way you were performing in order to get the job done?

· Step 3: Simulation Interview

o Pick a scenario uncovered during knowledge audit or from training docs

o For each major event, ask the following:

§ What actions, if any, would you take at this point in time?

§ What do you think is going on here? What is your assessment of the situation at this point in time?

§ What pieces of information led you to this situation assessment and these actions?

§ What errors would an inexperienced person be likely to make in this situation?

· Step 4: Compose a Cognitive Demands Summary Table

o | Difficult Cognitive Element | Cues & Strategies Used |

o | | |

o | Coordinating with right person… | Learn the presentation style |

Know the purposes of, and how to interpret (given a legend):

· Operational sequence diagrams

· Workload analysis diagrams

Slide Set #7

Why HRI is different from HCI

· Complex, dynamic control systems

· Autonomy and learning

· Real world environment

· Degraded operation

· Different roles

· Action at a distance

The HRI Awareness Base Case

Given one human and one robot working on a task together…

…HRI awareness is the understanding the human as of the

· Location

· Activities

· Status

· Surrounding of the robot

…the information that the robot has of

· The human’s commands necessary to direct its’ activities

· The constraints under which it must operate

Be able to state some examples of principles that apply to HRI (and why)

Principles that apply to HRI

1. Awareness

a. For one robot and one human, provide human-robot awareness of the robot’s location, surroundings, status, and activities, e.g.:

i. Provide location awareness via a map of where the robot is and has been

ii. Provide surroundings awareness via spatial information about the robot’s relationship to the environment, including automatic detection of obstacles

iii. Provide status awareness via automatic diagnosis-the ability for robots to self-inspect for damage-and report to the operator

iv. Provide activities awareness via video and sensor readouts

b. For Multiple Robots

i. Provide identity information about the robots

ii. Provide robot-robot awareness when needed

c. For multiple people:

i. Provide human-human awareness

2. Comprehensibility

a. Lower cognitive load by:

i. Providing fused sensor information

ii. Automatically presenting contextually-appropriate information

b. Provide help in choosing robot modality by determining the most appropriate level of robotic autonomy at any given time

c. Include:

i. A frame of reference

ii. Indicators of robot health/state including camera information

3. Consistency

a. If the operator is controlling multiple robots, the user interface for each should be consistent

b. If modes are used, be consistent in the functionality assigned to manipulators in each mode

4. Directness

a. Provide direct ways to accomplish tasks

i. Available alternatives should be visible

ii. The effect of actions on objects should be visible

5. Efficiency

a. Provide interfaces that support multiple robots in a single window, if possible

b. Minimize the use of multiple windows

c. Put frequently-accessed controls near one another to minimize hand movement

d. Do not include unnecessary information

e. Keep Fitts’ Law in mind when designing controls

6. Proximity

a. Both in controls and in the display, group like objects together

Slide Set #8

Definition of slips & mistakes

· Slips

o Users form an appropriate goal but don’t execute correctly

o Types of Slips:

§ Capture Errors

· Person starts doing intended activity but a familiar activity with same initial sequence “captures” attention

§ Description Errors

· Performing the correct action on the wrong object

· Most frequent when correct and incorrect objects are located near one another

§ Data-driven Errors

· Triggered by sensory data

§ Associative Activation Errors

· Triggered by internal thoughts and associations from prior experience

§ Loss-of-activation Errors

· Forgetting something

§ Mode Errors

· When a person executes a correct sequence but doesn’t realize the system is in the wrong mode

· Mistakes

o Users form the wrong goal as a result of conscious deliberation

Definitions of types of forcing functions

· Strong or physical constraints that make it easy to discover errors

· Types of Forcing Functions:

o Interlock

§ Forces operations to take place in the proper sequence

o Lock-in

§ Prevents user from prematurely stopping an action

o Lock-out

§ Prevents users from taking actions that are dangerous

How to design to minimize errors, including the role of structuring displays to support cognition

Characteristics of RPD/NDM settings

1. Time Pressure

2. High Stakes

3. Experienced Decision Makers

4. Inadequate Information (missing, ambiguous, or erroneous information)

5. Ill-defined goals

6. Poorly-defined Procedures

7. Cue Learning

8. Context (e.g., higher-level goals and stress)

9. Dynamic Conditions

10. Team Coordination

The differences between the Rational Choice and RPD models

· Rational Choice

o Thoroughly canvas a wide range of options

o Survey a full range of objectives

o Carefully weigh the costs, risks, and benefits of each option

o Intensively search for new information in evaluating options

o Assimilate all new information

o Re-examine the positive and negative consequences of each option

o Carefully plan to include contingencies if various risks occur

· RPD (Recognition-Primed Decision making)

o RPD is an outgrowth of Naturalistic Decision Making:

§ The study of how people use their experience to make decisions in field settings

o Definition of Decisions:

§ Choice points where multiple reasonable options are available

§ Regardless of whether the decision-maker actually considers multiple options

o This is what experienced decision-makers do:

§ Compare the situation to patterns or prototypes

§ Quickly evaluate courses of action by imagining how they will be carried out

§ Use the first workable option they can find

§ Imagine how the option will be carried out to spot weaknesses and find ways to avoid these

§ Generate other options if the first one doesn’t work

§ Remain poised to act rather than staying in analysis mode longer than necessary

o The RPD Model

§ Experience the situation in a changing context

§ Perceive situation as typical (a prototype or analogue)

§ Recognize and develop:

· Expectations

· Relevant Cues

· Plausible Goals

· Typical Action

§ Implement course of action

o How to do an RPD interview: First Pass

§ Interview by asking about the tough situations

§ Tape record interview

§ Treat each critical incident as a story

§ Have a checklist of items to probe, e.g.:

· Ways a person’s understanding changed during the episode

· Ways someone with less experience might have faltered

§ Use two analysts

· One to interview and the other to note-take and check whether all items have been covered

o How to do an RPD interview: Second Pass

§ Have interviewees go through story a second time to pin down what happened and when

§ Identify decision points

§ Ask whether interviewee thought about other courses of action

· If so, ask why the choice was made

· If not, ask what about the situation made it obvious

o Analyzing the data and putting it to use

§ Look at the decision strategy used

· Choosing from pre-selected options

· Comparative evaluation

· Novel option

· Recognition decision (singular evaluation)

§ Identify the information used to make decisions

· Where did the information come from?

· Can computer support make it easier to find the information and perform mental simulations?

The 4 “sources of power” used in RPD and their roles in the RPD process




Use of experiences to recognize key patterns that indicate the dynamics of a situation

Mental Simulation

Enables decision-makers to imagine how a situation will play out


Suggests parallels between the current situation and a previous one


Consolidation of experiences helps make them available to use in the future

The major components of the RPD model and how they interact:

Slide Set #9

The reasons to evaluate

· Why Perform Evaluations?

o Risk Reduction

§ Evaluation and early redesign reduces the risk that final UI design will not meet users’ needs

o Cost and Schedule Savings

§ Much less expensive to fix user interface problems during early project stages

§ Can avoid schedule slippage due to user interface changes late in the project

· When Should You Evaluate?

o Early design of an artifact to clarify design ideas

o Evaluation of a working prototype

o Refining or maintaining a product

o Exploring a new design concept


The major differences between the three approaches to evaluation (Major Evaluation Techniques)

1. Inspection: experts examine an interface

a. Example: heuristic evaluation

2. Empirical: users involved

a. Example: usability testing

3. Formal methods: analytical techniques

a. Example: Goals, Operators, Methods, Selection Rules

· Heuristic Evaluation

o an approach to evaluation in which knowledge of typical users is applied, often guided by heuristics, to identify usability problems

· How to do a heuristic evaluation

o Identify 3-5 evaluators (inspectors)

o Tailor heuristics, if needed, and discuss tailoring so all evaluators understand heuristic usage

o Each evaluator independently:

§ Examines an interface and judges its compliance with heuristics

§ Rates the severity of each problem found

o After independent inspections, evaluators meet to:

§ Discuss findings

§ Eliminate duplicate problems

§ Agree on ratings and summary findings

o Nielsen’s Heuristics

§ Provide visibility into system status

§ Provide a match between system and real world

§ Allow both user control and freedom

§ Follow standards to ensure consistency

§ Prevent Errors

§ Allow for recognition rather than recall

§ Provide flexibility and efficiency of use

§ Use an aesthetic and minimalist design

§ Help users recognize, diagnose and recover from errors

§ Provide help and documentation

Why heuristic evaluation is effective

· Power of method is in aggregating problems found

o Each inspector (“test users” in graph) may find 20%-50% of problems

o 5 Inspectors may find 75%-80% after aggregation

· Validated by research

Slide Set #10

What WIMP stands for

· Windows, Icons, Menus, Pointers

How to characterize interfaces in terms of input and output, devices and methods

1. Advanced Graphical

2. Web

3. Speech/Voice

4. Pen, Gesture, Touch

5. Mobile

6. Appliance

7. Multimodal

8. Shareable

9. Tangible

10. Wearable

11. Augmented and Mixed Reality

12. Robotic

How to apply principles of HCI to non-WIMP interfaces

The “research and design issues” of each of the non-WIMP interaction types from Sharp et al. Ch. 6

· Command Interfaces

o CLI–the most relevant principle is consistency. The method used for labeling/naming the commands should be chosen to be as consistent as possible.

· WIMP/GUI Interfaces

o Windows–the biggest issue is allowing users to easily flow from window to window. The design principles of Spacing, Grouping, and Simplicity should be used (to organize overview of all available/open windows)

o Menus – the most important principle is grouping: like commands should be grouped together and predicted to be in that menu.

o Icons – they should match what the icon will do (i.e., a magnifying glass for searching) and represent real-world mappings. Text can be used to disambiguate similar icons.

· Advanced Graphical Interfaces

o Multimedia – audio triggers imagination, diagrams trigger ideas, text expresses details.

o Virtual Reality and Virtual Environments – needs to be life-like, but can be abstract.

o Information Visualization – needs to be easy to comprehend.

· Web-Based Interfaces – where am I? What’s here? Where can I go? Directness. Website should be compatible with users, such as blind or deaf. Compatible with text-only screen. Compatible with hardware such as slow computers, small screen. Compatible with software, such as browser. Consider foreign people.

· Speech Interfaces – keep commands natural to human speech, easily recover from errors, since it is so error-prone. Understand vague commands. Voice response is male/female/serious/silly/etc…

· Pen, Gesture, and Touchscreen Interfaces – these are used as a form of input

· Appliance Interfaces – simplicity and visibility are paramount. Easy to see and understand when cooking bagel or making coffee. Status information (feedback should be highly visible).

· Mobile Interfaces – speed & efficiency are difficult to achieve compared to non-mobile devices. Space constraints. Make compatible for fat fingers. Apps are scaled down versions of original non-mobile apps. Cut back to only necessary controls and make more “hidden” layers.

· Multimodal Interfaces – idea is to allow people to experience more flexible, efficient, and expressive interface. Many different inputs/outputs can be used at same time. Need to recognize multiple aspects of human behavior instead of just touch or voice.

· Shareable Interfaces – directness. User’s need to be identified. Need to apply principles of collaborative interfaces.

· Tangible Interfaces – explicit mapping between mapping and effect is critical. Should be flexible and simple (i.e.: flexibility & simplicity)

· Augmented & Mixed Reality Interfaces – should be simple and align with the real world. Different for serious (military/medical) versus (games/learning).

· Wearable Interfaces – comfort and portability. Control. Reliable & accurate for heart monitors. Privacy. Hygiene. Social Acceptance. Unknowingly recorded conversations.

· Robotic Interfaces – interface will be expected to show some emotion equivalent to human/animal. Design to be human like (talking) versus robot like (buttons)

Slide Set #11

Multi-touch computing

Different technical approaches

Research approach of incremental experiments

Participatory Design

The definition of participatory design (PD)

· An approach to the assessment, design, and development of technological and organizational systems that places a premium on the active involvement of workplace practitioners (usually potential or current users of the system) in design and decision-making processes

Participatory design tenets (recognize these)

1. Respect the users of technology

a. …regardless of their status in the workplace, technical know-how, or access to the organization’s funding

b. View every participant in a PD project as an expert in what they do, as a stakeholder whose voice needs to be heard

2. Recognize that:

a. Workers are a prime source of innovation

b. Design ideas arise in collaboration with participants from diverse backgrounds

c. Technology is but one option in addressing emergent problems

3. View systems as networks of people, practices, and technology embedded in particular organizational contexts

4. Understand the organization and the relevant work on its own terms, in its own settings

5. Address problems that exist and arise in the workplace, articulated by or in collaboration with the affected parties, rather than attribute from the outside

6. Find concrete ways to improve the working lives of co-participants by, for example:

a. Reducing the tedium associated with work tasks

b. Co-designing new opportunities for exercising creativity

c. Increasing worker control over work content, measurement, and reporting

d. Helping workers communicate and organize across hierarchical lines within the organization and with peers elsewhere

7. Be conscious of one’s own role in PD processes

a. Try to be a “reflective practitioner”


· (Plastic Interface for Collaborative Technology Initiatives through Video Exploration)

o “Plastic” is used here mostly in the sense of “easily molded or changed”

· Non-software based rapid prototyping environment of common office objects

· Ideas is to involve users:

o In requirements refinement (if needed), or

o In brainstorming design (if requirements are concrete)

The basics of how to use PICTIVE

· How to use PICTIVE: Preparation

o Identify and invite a diverse set of stakeholders

§ Include users, designers, implementers

o Ask attendees to prepare materials in advance

§ Ask users to write up step-by-step scenarios of their most important or frequent work tasks

§ Ask designers to prepare a short presentation on requirements and issues that they have identified

§ Ask developers to prepare initial system components and material on the basics of the technical approach

o Organizers gather material and create any specialized props such as paper representations of icons

o Principle of reciprocal preparation: all groups have invested in the session

· How to use PICTIVE: Starting the Session

o Begin session with homework presentations

§ Requirements help ground subsequent discussion

§ Users’ scenarios inform designers and implementers

§ Implementers’ explanation of technical approach helps users to generate a mental model of what is possible

o Presentations provide reciprocal education

· How to use PICTIVE: Engaging in Design

o Design team introduces several issues and asks for the group to generate ideas for resolving them

§ Attendees use paper, pens, Post-it notes, scissors, representations of menus or dialog boxes

o Ideally, session proceeds as a group brainstorming session with no single person driving discussion

§ All participants are given a chance to show and explain their design ideas

o Videotape session for later analysis

§ Concentrate on reason why stakeholders suggest certain design approaches

The main difference between PD and UCD

· User-Centered Design according to Norman

o Make sure that the user:

§ Can figure out what to do

§ Knows what’s going on

o Follow 7 principles for making difficult tasks easier:

§ Use both knowledge in the world and knowledge in the head

§ Simplify the structure of tasks

§ Make things visible

§ Get the mappings right

§ Exploit the power of constraints

§ Design for error

§ When all else fails, standardize

o User Centered Design

§ Early focus on users and tasks

§ Empirical measurement

§ Iterative design

No comments: