CS 6750 Test 1
80 flashcards covering Norman's action cycle, affordances, cognitive load, needfinding, prototyping, and research ethics for HCI
What You'll Learn
Free flashcards for Human-Computer Interaction Test 1 topics: Gulf of Execution and Evaluation, slips vs. mistakes, direct manipulation, mental models, affordances, cognitive load, needfinding methods, design alternatives, prototyping, and research ethics. Ideal for graduate HCI students studying CS6750 at Georgia Tech.
Key Topics
- Gulf of Execution and Gulf of Evaluation: stages and design implications
- Slips vs. mistakes: causes, examples, and prevention strategies
- Direct manipulation: properties and when to apply it
- Affordances, mappings, constraints, and discoverability
- Cognitive load: types, reduction strategies, and distributed cognition
- Needfinding methods: interviews, observations, think-aloud, ethnography
- Design alternatives: brainstorming, chainstorming, personas
- Prototyping: paper, card, Wizard of Oz, wireframes, and fidelity trade-offs
- HCI design lifecycle and iterative design
- Research ethics: informed consent, IRB, social desirability bias, selection bias
Looking for more human computer interaction resources? Visit the Explore page to browse related decks or use the Create Your Own Deck flow to customize this set.
How to study this deck
Start with a quick skim of the questions, then launch study mode to flip cards until you can answer each prompt without hesitation. Revisit tricky cards using shuffle or reverse order, and schedule a follow-up review within 48 hours to reinforce retention.
Preview: CS 6750 Test 1
Question
What is the Gulf of Execution?
Answer
The distance between a user's goals and the actions required to achieve them. A wide gulf means users struggle to figure out what to do.
Question
What is the Gulf of Evaluation?
Answer
The distance between the system's output/state and the user's ability to interpret whether their goal was achieved. A wide gulf means users can't tell if their action worked.
Question
What are the 3 stages of the Gulf of Execution?
Answer
1) Goal formation (what do I want?), 2) Translating the goal into system terms (what can the system do?), 3) Executing the action (how do I do it?)
Question
What are the 3 stages of the Gulf of Evaluation?
Answer
1) Perceiving the system's state, 2) Interpreting the state, 3) Evaluating whether the goal was achieved
Question
What is a 'slip' in HCI?
Answer
An error where the user has the correct mental model/intention but executes the wrong action (e.g., pressing the wrong key). The user knows what to do but does something different.
Question
What is a 'mistake' in HCI?
Answer
An error where the user has an incorrect or incomplete mental model—they want the right outcome but choose the wrong action because they don't understand the system.
Question
How do you prevent slips vs. mistakes?
Answer
Slips: Use constraints and better mappings so the wrong action is harder to perform. Mistakes: Improve discoverability and representations so users develop a correct mental model.
Question
What is Direct Manipulation?
Answer
An interaction style where users interact directly with a visual representation of objects. Features: continuous representation, physical actions instead of syntax, rapid/incremental/reversible operations, immediate feedback.
Question
What are the key properties of Direct Manipulation interfaces?
Answer
1) Continuous representation of objects, 2) Physical actions instead of typed commands, 3) Rapid, incremental, reversible operations, 4) Immediate feedback on results
Question
What is a mental model?
Answer
The user's internal understanding of how a system works. Good design aligns the system's behavior with users' existing mental models to reduce learning and errors.
Question
What is a 'representation' in HCI?
Answer
The way a system displays or depicts its underlying state/content. Good representations make the task easier by aligning with the user's mental model and making important information perceptible.
Question
What are Norman's criteria for a good representation?
Answer
1) Makes relationships visible, 2) Captures the important distinctions, 3) Reveals natural constraints, 4) Allows inference, 5) Is complete but doesn't overload
Question
What is an 'affordance'?
Answer
A property (real or perceived) of an object that signals how it can be used. E.g., a door handle affords pulling; a button affords pressing.
Question
What is a 'mapping' in interface design?
Answer
The relationship between controls and their effects. Good mappings make it obvious which control does what (e.g., stove dials arranged spatially like the burners they control).
Question
What is 'perceptibility' as a design principle?
Answer
The user's ability to perceive the current state of the system. Interfaces should always keep users informed about what is going on through appropriate, timely feedback.
Question
What is the design principle of 'consistency'?
Answer
Using controls, layouts, and interactions uniformly—within an interface and across the broader design community—so lessons learned from one system transfer to another.
Question
What is the 'flexibility' principle?
Answer
Designing interfaces to support multiple ways of doing the same task to accommodate both novice and expert users (e.g., both right-click menus and keyboard shortcuts for cut/paste).
Question
What is the 'equity' principle in universal design?
Answer
The design should be useful to people with diverse abilities, providing the same means for all users (identical when possible, equivalent when not), without stigmatizing any users.
Question
What is 'cognitive load' in HCI?
Answer
The mental effort required to use an interface. Good design minimizes unnecessary cognitive load so users can focus on their task rather than the interface.
Question
What are the 5 tips for reducing cognitive load?
Answer
1) Use multiple modalities, 2) Let modalities complement (not compete with) each other, 3) Give user control of pace, 4) Emphasize essential content, minimize clutter, 5) Offload tasks from user onto the interface
Question
What are the three types of human perception most used in UI design?
Answer
1) Visual (sight), 2) Auditory (sound/hearing), 3) Haptic (touch/force feedback)
Question
What is 'distributed cognition'?
Answer
A framework viewing cognition as spread across individuals, artifacts, and the environment—not confined to one person's mind. Interfaces can serve as cognitive tools (e.g., a cockpit's speed cards augmenting memory).
Question
What cognitive roles do interfaces play in distributed cognition?
Answer
Primarily augmenting memory (long-term and short-term), but also supporting inference, computation, and coordination across people and artifacts.
Question
What is 'invisible design' (design becoming invisible)?
Answer
When an interface is so well-designed that users stop thinking about it and focus entirely on the task. Good design goes unnoticed; learning the interface becomes automatic.
Question
What is 'task invisibility by learning'?
Answer
When frequent use of an interface causes the interaction steps to become automatic/habitual, shifting the user's attention entirely to the task goal rather than the interface itself.
Question
What is 'needfinding'?
Answer
The first stage of the design lifecycle: gathering data to understand who users are, what they do, and what they need—before designing. The biggest mistake is jumping to design without understanding the user.
Question
What data-gathering methods are used in needfinding?
Answer
Observations (watching users in context), interviews (1-on-1), focus groups, surveys, think-aloud protocols, diary studies, and ethnography/apprenticeship for complex tasks.
Question
What is a 'think-aloud protocol'?
Answer
A method where users verbalize their thoughts while performing a task, revealing their mental model in real-time. Useful but may alter behavior by making users more deliberate.
Question
What is 'social desirability bias' in research?
Answer
The tendency for participants to give answers they think the researcher wants to hear (e.g., saying they'd use an app to avoid seeming unsupportive). Avoided by neutral question framing.
Question
What are the 5 tips for conducting effective interviews?
Answer
1) Focus on the 6 W's (who, what, where, when, why, how), 2) Avoid bias in phrasing, 3) Listen—let the participant talk, 4) Organize the interview (intro, questions, summary), 5) Practice beforehand
Question
Why avoid yes/no questions in interviews?
Answer
They limit response depth and don't reveal the participant's actual thinking. Open-ended questions elicit richer data and uncover perspectives the researcher may not have anticipated.
Question
What is the difference between interviews and focus groups?
Answer
Interviews are 1-on-1 and allow deep individual responses. Focus groups involve multiple participants simultaneously and can surface diverse perspectives but risk 'groupthink' and overly convergent responses.
Question
What is 'ethnography' in HCI research?
Answer
A method where the researcher immerses themselves in users' community or work environment—becoming a participant—to deeply understand complex tasks that can't be grasped through brief observation.
Question
What are 'design alternatives'?
Answer
Multiple distinct approaches to solving a design problem, generated before committing to one solution. Generating alternatives avoids fixating on the first idea and surfaces trade-offs.
Question
What is 'brainstorming' in HCI design?
Answer
A generative ideation technique for rapidly producing many design alternatives. Key rules: defer judgment, go for quantity, encourage wild ideas, build on others' ideas.
Question
What is 'chainstorming'?
Answer
A variant of brainstorming where each idea prompts the next in a chain, helping overcome blocks by following a thread of connected ideas rather than generating independently.
Question
What is a 'persona' in HCI design?
Answer
A fictional but data-grounded character representing a target user type, including their goals, context, behaviors, and frustrations. Used to keep design focused on real user needs.
Question
What is a 'user task' vs. a 'user goal'?
Answer
A task is the specific action/activity a user performs (e.g., 'click save'). A goal is the underlying outcome they want (e.g., 'not lose my work'). Good design addresses goals, not just tasks.
Question
What is a prototype in HCI?
Answer
An early, incomplete version of a design used to test ideas and gather feedback before committing to full development. Ranges from low-fidelity (paper) to high-fidelity (working software).
Question
What is a low-fidelity prototype?
Answer
A rough, inexpensive representation of a design (e.g., paper sketches, card prototypes, verbal descriptions) used to explore ideas early. Easy to revise, hides superficial details.
Question
What is a high-fidelity prototype?
Answer
A more complete, polished prototype (e.g., wireframes, working software) that closely resembles the final product. Better for testing look-and-feel and realistic interactions.
Question
What is a 'Wizard of Oz' prototype?
Answer
A prototype where a human (the 'wizard') simulates system behavior behind the scenes while users believe they're interacting with an automated system. Useful for testing AI/voice interfaces cheaply.
Question
What is a 'card prototype'?
Answer
A low-fidelity prototype using physical cards to represent screens/states. Users can navigate by choosing/flipping cards, simulating interaction flow without coding.
Question
What is a 'wireframe'?
Answer
A mid-fidelity schematic of an interface layout showing structure and navigation without detailed visual design. Useful for distributing to remote users and testing screen flow.
Question
What does a prototype 'disguise superficial details'?
Answer
Low-fidelity prototypes prevent users from fixating on fonts, colors, and aesthetics, keeping feedback focused on structure, functionality, and interaction patterns.
Question
What does 'Houde & Hill' say prototypes prototype?
Answer
Prototypes can address: 1) Role (what the artifact does for the user), 2) Look and feel (user experience/aesthetics), 3) Implementation (how it works technically), or combinations of these.
Question
What is the HCI design lifecycle?
Answer
An iterative process: 1) Needfinding/requirements gathering, 2) Design alternatives, 3) Prototyping, 4) Evaluation/user testing → repeat until design meets user needs.
Question
What is a 'cognitive task analysis' (CTA)?
Answer
A method of identifying the knowledge and mental steps required for a user to complete a task—including what the user must know and where that knowledge comes from.
Question
What is a 'heuristic evaluation'?
Answer
An expert review of an interface against a set of established usability principles (heuristics) to identify design violations—without user testing.
Question
What are Nielsen's 10 Usability Heuristics (key ones)?
Answer
1) Visibility of system status, 2) Match between system and real world, 3) User control & freedom, 4) Consistency & standards, 5) Error prevention, 6) Recognition over recall, 7) Flexibility & efficiency, 8) Aesthetic & minimal design, 9) Help users recognize/recover from errors, 10) Help & documentation
Question
What is 'error prevention' vs. 'error recovery'?
Answer
Prevention: Design the interface to make errors hard to commit (constraints, good mappings). Recovery: When errors do occur, make them easy to undo or recover from with clear feedback.
Question
What is 'recognition over recall'?
Answer
A usability principle stating interfaces should make options visible/recognizable rather than requiring users to remember them from memory (e.g., menus vs. command-line interfaces).
Question
What is 'informed consent' in HCI research ethics?
Answer
Participants must be told what the study involves, agree voluntarily, and understand they can withdraw. Consent should be specific to the study, time-limited, and free from coercion.
Question
What were the ethical problems with Facebook's 2014 emotional contagion study?
Answer
1) Users weren't specifically notified they were in a study, 2) They couldn't opt out without deactivating accounts, 3) Agreeing to terms of service ≠ genuine informed consent to a specific experiment.
Question
What is an IRB (Institutional Review Board)?
Answer
A committee that reviews research involving human subjects at federally-funded institutions to ensure ethical standards are met (informed consent, participant rights, minimizing harm).
Question
What are the core ethical principles for human subjects research?
Answer
1) Respect for persons (autonomy/informed consent), 2) Beneficence (maximize benefits, minimize harm), 3) Justice (fair distribution of research burdens and benefits)
Question
What is the difference between 'independent' and 'dependent' variables?
Answer
Independent variable: what the researcher manipulates (the cause). Dependent variable: what is measured as an outcome (the effect). E.g., IV = interface design; DV = task completion time.
Question
What are the 4 levels of measurement?
Answer
1) Nominal (categories, no order), 2) Ordinal (ordered categories, no equal intervals), 3) Interval (equal intervals, no true zero), 4) Ratio (equal intervals + true zero, e.g., reaction time)
Question
What is a null hypothesis?
Answer
The default assumption that there is no effect or difference between conditions. The goal of statistical testing is to determine if evidence is strong enough to reject it.
Question
What is a t-test used for?
Answer
To compare means between two groups (or a group to a known value) to determine if the difference is statistically significant. Requires interval/ratio data and roughly normal distributions.
Question
What is an ANOVA?
Answer
Analysis of Variance—a statistical test for comparing means across 3 or more groups simultaneously. An extension of the t-test that avoids inflating Type I error.
Question
What is a 'participant' view of the user?
Answer
Considering the full context surrounding a user's activity—physical environment, cognitive state, social situation, emotional state—not just their interaction with the interface.
Question
What is 'context' in HCI and why does it matter?
Answer
Context is the situation surrounding an interaction (e.g., walking, driving, distracted). Context changes cognitive and physical resources available and constrains how interfaces should be designed.
Question
What is the 'processor' model vs. 'participant' model of the user?
Answer
Processor: Views user as an information-processing machine. Participant: Views user as a full person situated in a rich social/physical/cultural context. HCI increasingly favors the participant view.
Question
What is 'universal design'?
Answer
A design philosophy advocating that products and environments should be usable by all people, to the greatest extent possible, without adaptation or specialized design.
Question
What are the 7 principles of universal design?
Answer
1) Equitable use, 2) Flexibility in use, 3) Simple & intuitive use, 4) Perceptible information, 5) Tolerance for error, 6) Low physical effort, 7) Size & space for approach & use
Question
What is 'Norman's Action Cycle'?
Answer
A 7-stage model: 1) Goal, 2) Plan, 3) Specify, 4) Perform, 5) Perceive, 6) Interpret, 7) Compare. The first four cross the gulf of execution; the last three cross the gulf of evaluation.
Question
What are 'constraints' in interface design?
Answer
Design features that limit user actions, making incorrect actions harder or impossible to perform. They prevent slips and guide users toward correct behavior (e.g., grayed-out menus, physical fit).
Question
What is 'discoverability'?
Answer
The degree to which users can figure out how to use an interface without prior instruction—by exploring it. High discoverability means affordances and mappings are clear enough to guide new users.
Question
What is a 'feedback cycle' in HCI?
Answer
The loop through which a user takes an action, the system responds, and the user perceives and interprets that response to determine if their goal was met. Good design creates tight, clear feedback cycles.
Question
What is 'human-centered design (HCD)' and what critiques exist?
Answer
HCD places user needs at the center of design. Critique (Norman's paper): Over-emphasis on human needs can neglect system constraints, sometimes making designs less sustainable or efficient at a broader scale.
Question
What is the difference between a 'survey' and an 'interview' for research?
Answer
Surveys are scalable, structured, and good for quantitative/yes-no data from many participants. Interviews are 1-on-1, open-ended, and good for deep qualitative insights but harder to scale.
Question
What is 'social desirability bias' and how is it avoided?
Answer
Participants answer how they think the researcher wants, not honestly. Avoided by neutral phrasing, not revealing the researcher's hypothesis, and assuring participants there are no right/wrong answers.
Question
What is 'selection bias' in research recruitment?
Answer
When the recruited sample doesn't represent the target population—e.g., recruiting only tech-savvy volunteers skews results. Consent procedures requiring effort can introduce selection bias.
Question
What is 'think-aloud' and when is it most valuable?
Answer
Asking users to verbalize thoughts while performing tasks, revealing real-time cognitive processes. Most valuable when you want to understand *why* users make decisions, not just *what* they do.
Question
What are 'design heuristics' and who created the most famous set?
Answer
Principles used to evaluate interfaces for usability problems. Jakob Nielsen created the most widely-used set of 10 heuristics for interface design evaluation.
Question
What is the 'iterative design' process?
Answer
A cyclical process of designing, prototyping, evaluating, and refining—repeated until the design adequately meets user needs. Avoids committing to a final design too early.
Question
What distinguishes 'verbal prototype' from 'paper prototype'?
Answer
Verbal prototype: the designer describes the interface in words to get early reactions. Paper prototype: a hand-drawn, tangible mockup users can interact with by pointing or manipulating cards.
Question
What is the 'Norman door' problem?
Answer
A door whose design signals the wrong affordance (e.g., a pull handle on a push door). It's an example of poor mapping between affordance and actual function—the user's action is predictably wrong.