Usability Principles, Execution vs. Evaluation, and Interactive System Analysis

Policy on Academic Integrity and AI Tool Usage

• Full disclosure of AI tool usage is strictly required. Non-disclosure of AI tool usage is significantly worse than transparent disclosure.

• Acceptable uses of AI tools include:

  • Providing formatting support
  • Proofreading written work
  • Organizing personal ideas into a more coherent narrative structure

• Unacceptable uses of AI tools include:

  • Inputting prompts into AI generators to complete assignments on the user's behalf.

• Submissions are audited using detection software. If an automated AI detection score exceeds $50\%$, the submission undergoes detailed manual inspection to identify telltale signs of synthetic generation.

• Final academic decisions are not based solely on an automated detection percentage, but on a holistic evaluation of the submitted work.

Usability Analysis Assignment Specifications

Overview and Core Objectives

• The central objective of the assignment is to evaluate and predict the usability and user experience of interactive systems.
• Critiques must extend beyond simple subjective opinions (such as stating "I like this" or "I dislike this") and must strictly align with established human-computer interaction (HCI) literature and design principles.
• The assignment emphasizes authentic, hands-on critical analysis rather than automated content generation.

Step-by-Step Assignment Instructions

1. Select an Interactive Experience: Choose a single specific experience interacting with an interactive system. The system can be either a physical interface or a digital interface, provided it is interactive. It does not need to be a digital product, nor should it rely on watching a recording of someone else interacting with the system.
2. Define Goal and Map to Action Cycle: Formulate the explicit goal attempted during the experience. Describe the progression of the interaction using the stages of Norman's Action Cycle as the explanatory framework.
3. Evaluate Design Principles (User Perspective): Analyze the experience from the perspective of the user across six designated design principles. Detail what succeeded or failed, justifying why each element fits the specified principle.
4. Propose Redesign Strategy (Designer Perspective): Adopt the perspective of the system designer to analyze underlying causes. Explicitly document:
   • Aspects of the design that should remain unchanged and why.
   • Aspects of the design that must be modified and why.
   • How every decision addresses each of the core design principles.

Required Usability Principles

The analysis must individually address six foundational principles:

• Visibility / Discoverability: Whether interface options and controls can be easily located and identified.
• Feedback: Information sent back to the user to confirm what action was taken and what result was achieved.
• Conceptual Model / System Model: The underlying structure and coherent mental representation of how the system operates.
• Affordances and Signifiers: The physical or digital action possibilities offered by an object (affordances) and the explicit indicators that signal those possibilities (signifiers).
• Mapping: The spatial, physical, or logical relationship between controls and their corresponding real-world effects.
• Constraints: Physical, logical, semantic, or cultural boundaries that restrict the range of possible actions to prevent errors.

Format and Logistics

• Word Count Limit: Maximum $1,000$ words.
• Structural Flexibility: Submissions may use headings structured around the stages of the action cycle, subheadings organized by usability principles, narrative prose, or bulleted lists.
• Authenticity Requirement: Assignments must reflect direct, personal interaction with the selected system to retain granular narrative details.
• Submission Timeline: Due approximately $2$ weeks from assignment release.

Execution versus Evaluation in Human-Computer Interaction

Core Interaction Concepts

Interaction with interactive systems is divided into two distinct components:

• Execution: The operational phase where a user attempts to enact a goal ("How do I make it happen?").
• Evaluation: The interpretative phase where a user assesses the system's updated state ("What happened, and does it match my intent?").

The Gulf of Execution

• The Gulf of Execution represents the gap between a user's internal goal and the actions made available by the interface.
• Key diagnostic questions to identify execution gaps include:
  • What is the user attempting to accomplish?
  • Can the user discover what actions are currently possible?
• An execution breakdown occurs when a user states: "I know what I want to do, but I do not know how to do it or where to go."

Causes and Real-World Examples of Execution Failure

• Unlabeled or Cryptic Icons:
  • Attempting to save a document when presented with unfamiliar icons creates a severe discoverability barrier.
  • Long-established icons (e.g., a recycling bin modeled after a physical trash bin) successfully map to pre-existing real-world mental models.
  • Modern arbitrary icons or emojis often fail to map cleanly, leaving users forced to guess whether an action will delete data, relocate a file, or perform an unintended operation.
• Wayfinding and Multi-Modal Transit Systems:
  • Navigating complex transportation hubs (e.g., Penn Station in New York) creates high cognitive load and execution failure for new users when signage fails to distinguish between different transit networks (regional rail, streetcars, buses).
  • Regular commuters navigate these hubs efficiently through repeated exposure, whereas first-time travelers experience severe confusion without explicit step-by-step guidance.
• Paper and Digital Forms:
  • Forms that fail to visually distinguish required fields from optional fields create uncertainty during execution.
• Norman Doors:
  • Doors lacking intuitive push or pull mechanisms present a classic physical execution breakdown.

Design Strategies to Bridge the Gulf of Execution

• Combine explicit text labels with icons rather than relying entirely on visual imagery.
• Ensure essential controls remain clearly visible.
• Establish intuitive default settings.
• Implement system constraints to narrow down valid options and prevent illegal actions.
• Utilize secondary visual cues (e.g., hover-over tooltips) to explain control functionality without sacrificing visual aesthetics.
• Reconcile aesthetics and functionality through creative design rather than treating them as mutually exclusive trade-offs.

User Diversity, Mental Models, and System Image

Designing for Novice versus Expert Demographics

• Users cannot be treated as a single monolithic block. Experience levels, goals, and communication styles vary significantly across user groups and generational cohorts (e.g., Zoomers versus older generations).
• Systems that force a single uniform workflow create sub-optimal user experiences:
  • Expert / Frequent Users: Prioritize speed and expediency (e.g., requiring a direct, single-click process to add money to a transit Key Card via phone or kiosk).
  • Novice / First-Time Users: Require step-by-step guided walkthroughs to navigate unfamiliar tasks.
• Interfaces should offer customized, differentiated pathways tailored to different experience levels rather than defaulting to a single static interface.

System Image, User Models, and Metaphors

• User Model: The internal mental model and expectations a user holds regarding how a system functions.
• System Image: The visual, physical, and functional reality created by the designer (including controls, layout, and documentation).
• Effective design requires the system image to directly support and reinforce the user model.
• Desktop Metaphor: Dragging a file icon into a trash bin to delete it aligns with real-world physical metaphors established over $20$ to $30$ years (originating in early computer interfaces such as $386$ systems).
• Universal Cultural Metaphors: Visual signifiers such as a stylized person symbol intuitively communicate the location of a restroom.

Consequences of Mismatched Mental Models

When a mismatch occurs between the system image and the user model, the interaction results in:

• User surprise
• Hesitation
• Operational errors
• A dependence on external instructional manuals

Practical Evaluation Framework

• Assessing interface usability requires identifying real-world execution designs:
  • Good Execution Design: Flat push plates on doors designed for pushing; reverse graspable handles on doors designed for pulling; intuitive trash bin metaphors for file deletion.
  • Poor Execution Design: Unlabeled icon bars; doors requiring pulling that feature push-plate hardware; wayfinding signage with ambiguous directionality.