ASU CSE 463 with Seifi Spring 2024 Final Exam Quizlet

Types of Data

Types of Data

Categorial: Names, no order or numerical relationship. Ex: Yes/No

Ordinal: Similar to categorical but there is an order between categories. Ex: low, medium, high

Interval/Numerical: Similar to ordinal but the interval between values it equally proportional. Ex: Hourly income $10, $15, $20

T Test

Test to determine the probability that your samples from A and B testing are from one distribution (Null Hypothesis)

When To Reject Null Hypothesis

When Probability is less than 5%. This means that the two sets of samples are NOT normally distributed.

When To Accept Null Hypothesis

When Probability is more than 5%. This means that the two sets of samples ARE normally distributed.

Parametric Test

Make assumptions about the underlying distribution of data to provide more statistical power. Use when data is interval AND assumptions are met. Ex: T Test

Non Parametric Tests

Do NOT make assumptions about data distribution but also less powerful statistically wise. Use when data is categorial or ordinal OR some assumptions are rejected

T Sample Assumptions

-Data points are independent
-Distribution of data in each condition is approximately normal
-There is homogeneity of variance

Assumption of Normality

Your dependent variable is normally distributed

Participant Based Evaluations

-Empirical data showing how users interact with systems
-Thinking Aloud Testing
-Wizard of Oz
-AB Testing

Expert Based Evaluations

Not directly involving users. Faster and cheaper than user evaluation, but require practice and focus on usability. They also dont replace user evaluation.

10 Usability Heuristics

-Visibility of System Status
-Match Between the System and the Real World
-User Control and Freedom
-Consistency and Standards
-Error Prevention
-Recognition Rather Than Recall
-Flexibility and Efficiency of Use
-Aesthetics and Minimalist Design
-Help Users Recognize, Diagnose, And Recover From Errors
-Help and Documentation

Visibility of System

The system should always keep the user informed about whats going on.

Match Between the System and the Real World

The language should speak the users language with words that are familiar to the user, no jargon.

User Control and Freedom

Users often choose systems functions by mistake and will need clearly marked emergency exit to leave the unwanted state without having to go through an extended dialogue.

Consistency and Standards

Users should not have to wonder whether different words, situations, or actions mean the same thing. Use colours consistently. Follow platform conventions.

Error Prevention

Design that eliminates error prone conditions or check for them and present the user with a confirmation option before they commit to the action.

Recognition Rather Than Recall

Minimize users memory load by making objects, actions, and options visible. The user should not have to remember info from one part of the interface to another.

Flexibility and Efficiency of Use

Systems should include accelerators which are unseen to the novice user but allow the expert to navigate faster with frequent actions.

Aesthetics and Minimalist Design

Do not include info that is irrelevant or rarely needed.

Help Users Recognize, Diagnose, And Recover From Errors

Error messages should be expressed in plain language, no codes, precisely indicate the problems, and constructively suggest a solution.

Help and Documentation

Systems should be used without documentation, but if needed documentation should be easy to search, have concrete steps, and not be too large.

Severity of a Problem (Heuristic Evaluation)

0: I dont agree that this is a usability problem at all
1: Cosmetic problem
2: Minor usability problem
3: Major usability problem
4: Usability catastrophe

Criteria of Severity Rating (Heuristic Evaluation)

-Frequency
-Impact
-Persistence

Cognitive Walkthrough

-Select a goal that users might want to achieve with the system
-Determine the actions to accomplish the goals using the system
-For each step, answer four questions
-Provide an overview of the results and an elaborated explanation

Cognitive Walkthrough Questions

-Will the user understand this step is needed to reach their goal?
-Will the user see the control to perform the action?
-Will the user recognize the control as the right one?
-Will the user understand the feedback?

Pros and Cons of Cognitive Walkthrough Summary

Pros:
-Systematic, chance of missing an issue is low
-Actionable, the questions help identify the reason for the issue and how to solve it

Cons:
-Very involved, hard to use for a system with many functionalities
-Sometimes hard to put yourself in the user's shoes

Social Computing

Computing with a significant social layer. Covers a wide range of topics from groupware to social media.

Values of Social Computing

-Breaking constraints
-Efficiency of processing
-Improving quality of outcomes
-Human powered computing

Two Major Threads in Social Computing

-Supporting people through technology (Groupware)

-Improving tech and computing through human input (Crowdsourcing and Human Powered Computation)

Groupware

Generic term for interactive systems that are designed for use by collaborative work groups.

Two Considerations for Groupware

-Awareness Features: Individuals working together need to be able to gain some level of shared knowledge about each others activities

-Tailorability or Adaption: The way technology is used by a team may be completely different from how the designers intended it.

Crowdsourcing

The act of taking a job traditionally performed by a designated agent and outsourcing it to an undefined, generally large group of people in the form of an open call. Combining responses from multiple users results in an accurate or desirable response.

Pros and Cons of Crowdsourcing

Pros:
-Fast data collection
-Cost effective

Cons:
-No control over the environment
-Less control over data quality

When Is Crowdsourced Data Valid?

-Simple choice based tasks
-Perceptual tasks
-User impression ratings
-Testing a new interaction

When Is Crowdsourced Data NOT Valid

-Cognitively demanding tasks
-Collecting rich qualitative data

Human Powered Computation

Tasks that are easy for people but difficult for algorithms are relayed to human crowd. Can be used to verify human user and to train new algorithms.

Challenges of AI

-Unpredictability, AI makes mistakes, sometimes unexpectedly
-Opaqueness: User has difficulty understanding how system works
-Evolution: AI behavior changes overtime

Models of AI Interaction

-Automate
-Augment

Automate

Take action on users behalf

Augment

Provide options or additional information via prompt, organize, and annotate

Desirable Characteristics for Human AI Collaboration

-Complementary Performance: Leverage strengths of both AI and Humans to achieve better outcomes

-Alignment: AI should behave in ways that humans expect their teammates to behave

-Cooperative Interaction: The interaction human and AI Should work in a way that's both intuitive for the human, and compatible to model characteristics.

-Reasonable Task Allocation: Assign tasks that the model can perform.

Explicit vs Implicit Interactions

Explicit: Users directly select an action or command

-Implicit: The system responds to or collects data on user actions in the background

WIMP Style Interfaces

-Windows
-Icons
-Menus
-Pointers

Windows

Shared graphical display resources among multiple applications.

Icons

Based on three main kinds of mapping:
Direct, Metaphorical, and Conventional

Direct Mapping

Image of what the object is trying to represent.

Metaphorical

Transferring properties from a real to a digital world element.

Conventional

More or less arbitrary, becomes accepted overtime.

Menus

Organize and store the commands, contains items which are organized by menu topics. Cascading vs Pop Up Menus.

Pointers

The means by which users interact with the interface. Ex: Mouse, joystick, VR controllers.

Direct Manipulation

Interaction paradigm used in WIMP style interfaces. Objects on the screen are directly manipulated. Three Principles: Visibility of objects and actions, physical actions, rapid and incremental and reversible actions.

Virtual Reality

Replaces vision entirely with a virtual environment. Navigate and interact with objects in seemingly real or physical ways.

Virtual Reality Illusions

-Place or illusion or presence
-Embodiment illusion
-Plausibility illusion

Place Illusion or Presence

The extent to which an individual experiences the virtual setting as the one in which they are consciously present.

Embodiment Illusion

The user believes that the body they have in the virtual world belongs to them.

Plausibility Illusion

The user believes the virtual events are actually happening.

Augmented Reality

Real objects are enhanced by computer generated perceptual information in the real world.

Mixed Reality

More ambiguous, several definitions. Somewhere between VR and AR (Augmented Reality)

Extended Realities

Catch all term for VR, AR, and MR (Mixed Reality)

Interactions in VR

-Locomotion: Moving in VR space
-Manipulation: Selecting, grasping, and changing objects

Locomotion in VR

-Stationary/Room Scale
-Redirected Walking
-Joystick
-Teleportation
-Gesture/Movement Based
-Specialized Hardwares

Stationary or Room Scale

VR and physical space match 1 to 1

Redirected Walking

Fooling peoples visual and body perception. Visual feedback is slightly rotated so that the user thinks they are walking straight but in reality they are turning.

Joystick

Use joystick controls to move to different locations.

Teleportation

The user selects a location at a distance and is moved to the new location with a button click.

Gesture or Movement Based

Users replicate some aspect of movement but they are stationary in the physical world.

Specialized Hardware

Users stand or move on a physical platform that moves as the VR needs.

Reality Based Interactions

Replicate real world movement and interactions closely. Ex: Room scale or stationary locomotion.

Illusory Interactions

Rely on fooling human perceptual system to convey a realistic interaction. Ex: Redirected Walking

Beyond Real or Magical Interactions

Give users superpowers to facilitate VR interactions. Ex: Teleporting, Movement or Gesture Based, Scaling the Space or User Movement.

Evaluating Locomotion Techniques

-Accuracy in relation to the desired target
-Ease of learning for new users
-Ease of user over time
-Speed with appropriate velocity, feeling, and control
-Spatial awareness during and after travel
-Info gathering during travel
-Presence or sense of being there with illusion
-Comfort including VR sickness and required energy.

VR Sickness

Physical discomfort that comes as a result of using VR.

Vergence Accommodation Conflict

Vergence: Independent inward/outward rotation of eyes to fixate on objects and see them as a single object.

Accommodation: Eyes focusing mechanism to produce a sharp image.

Visual Vestibular Mismatch

Cause of motion sickness where your body thinks its stationary but your eyes tell you that you are moving.

Interacting with Nearby Ojects

Interactions can closely resemble real world actions.

Ray Casting

Selecting and manipulating objects at distance is desirable. To do this ray casting is used where a ray of light is pointed at the target by the user and the user confirms the selection.

Go Go Technique

Close to body: Physical and VR hand location is mapped 1 to 1
After a Threshold: VR hand location is scaled.

Ninja Hands

When many objects are in the same scene, using many hands to improve target selection.

World in Miniature

Manipulating far objects is hard. A similar technique exists for locomotion.

Key Areas of Accessibility for VR

-Content Accessibility
-Accessible Input Techniques and Devices
-Inclusive Representations

Content Accessibility

Tools for making content available, like colour blind modes.

Accessible Input

More diverse input techniques like speech, gaze, use more sensing techniques.

Inclusive Representation

Allow for a diverse set of avatars.

Challenges of Touchscreens

-Touchscreens can lead to gulf of execution
-Is it a touchscreen?
-What can it sense?

Gestures

-A form of nonverbal communication
-A movement of part of the body, especially a hand or the head.
-Used in place of or to augment speech.

Gesture Recognition

-Look at patterns from the sensor to recognize a set of discrete poses or gestures.
-Interpreting human gestures via algorithms
-Map gestures to commands for devices and applications.

Challenges for Gesture Interfaces

-Lack of a universal gesture language
-Different users make gestures differently, causing difficulty in identifying motions
-Engineering problems such as impact of light and noise on the sensor performance
-Can be hard to add new gestures as they impact performance of previous ones
-Sometimes need calibration and training per user

Hand Tracking

-Capturing continuous motion rather than a gesture
-Hand Tracking is less problematic for users than gesture based systems since individual poses do not activate commands

Full Body Gesture Recognition

Similar to hand gestures but using wearables.

Full Body Motion Tracking

Only relies on headset and controllers, no external tracking.

Speech Based Interfaces

Use speech for interfaces.

Speech Recognition: The computer understanding what the customer is saying.

Speech Production (or Synthesis): The computer talking to the customer.

Pros of Speech Based Interfaces

-People can talk at ~90 words per minute (high speed)
-"Virtually Unlimited" set of commands
-Freedom for other body parts
-Used by people with visual impairments
-e.g., speech recognition word processors, page scanners, web readers, and home control systems

Cons of Speech Based Interfaces

-Gulf of Execution, hard to know available commands

-Gulf of evaluation speech recognition far from perfect.

Challenges of Gaze Based Interactions

-Error prone
-Require training
-Require careful design of interactions/commands
-Clever designs can leverage gaze effectively

Haptics

Anything relating to the sense of touch.

Tactile

Via skin

Kinesthetic

Via muscles

Haptic Interface

-Senses a physical quantity from the user (motion, force, etc)
-Physically acts on the user via an actuator
-Connects sensing to acting with fast processing.