18 Understanding and Conceptualizing Interaction (Part 1)
18.1 The Problem Space
The Problem Space refers to the set of issues, needs, goals, and constraints that motivate the development of a system.
Before proposing a solution, designers must define the problem space by answering critical questions:
- Creation & Audience: What are we trying to create, and for whom?
- Assumptions: What assumptions are being made about user goals and behaviors?
- Effectiveness: Will the solution achieve goals within the specific constraints of the context?
- Integration: Are there existing systems or workarounds that the new design should improve upon?
18.2 Framework for Analyzing the Problem Space
There is a structured framework for critically examining the problem space and uncovering hidden assumptions.
Before building, designers must understand the current landscape of the user’s struggle. This includes identifying existing limitations. Root Cause Analysis can be used to determine the underlying cause of these problems.
Because a system doesn’t exist in a vacuum, designers must consider how it fits into the user’s existing world. They should consider Workflow Integration, Task Efficiency, and how it enhances the user experience.
This framework promotes Reflective Thinking and design Alignment to real-world needs and constraints.
18.3 Transitioning from Problem Space to Design Space
While the Problem Space defines the “What” and “Why,” the Design Space explores the “How”.
- Problem Space: Focuses on users, their tasks, goals, and existing constraints.
- Design Space: The realm of possible solutions, interface types, and system behaviors.
To move into the design space, designers must balance 4 pillars:
- Translation
- Identification
- Exploration
- Alignment
18.4 The Conceptual Model
The conceptual model is a high-level description of how a system is organized and operates. It is done to provide a simplified, understandable representation of complex system logic. It allows users to predict outcomes of their actions before they take them. There are 3 components of a conceptual model
- User Actions: Defines what the user can do
- System Responses: How the system behaves when an action is taken.
- Interface Representation: How concepts are visualized. This helps the user form a correct mental model.
The primary goal of the conceptual model is to ensure the user’s mental model matches the system reality. Leading to:
- Predictability: If the model is clear, the user isn’t surprised by the system’s behavior.
- Error Reduction: This reduces errors and mistakes in the system usage.
- Increased Confidence: Users feel in control of the technology.
A good conceptual model is important because it leads to:
- Consistency: Provides a framework so that new features feel like they “belong” to the same system.
- Design Guidance: Helps designers choose the right interface elements (buttons, sliders, or menus) based on the underlying model.
- Alignment: Ensures that what the user expects to happen is exactly what actually happens.
- Crucial for Success: Especially vital when designing complex systems where technical logic may not be inherently obvious to the end-user.
A successful conceptual model is the result of aligning these four elements:
- User Tasks (The Journey)
- System Support (The Guide)
- Metaphors (The Map)
- Interaction Styles (The Vehicle)
18.4.1 System Support, Interface Metaphors, and Interaction Styles
The system must guide the user through the process using familiar concepts. Two techniques are:
- System Support: Providing the right functionality to the user at the right time. This includes proactive guidance and clear feedback.
- Interface Metaphors: Leveraging real-world analogies to flatten the learning curve.
Designers must select the “language” of the interaction based on the user’s context and needs. There are 3 common interaction styles:
- Direct Manipulation: Dragging an icon to the trash (immediate and visual).
- Conversational: Asking a voice assistant to set an alarm (natural language).
- Menu-Based: Selecting options from a structured list (low memory load).
18.5 First Steps in Formulating a Conceptual Model
- Understanding User Tasks
- System Support & Interface Metaphors
- Interaction Modes and Styles
18.5.1 The Designer’s Foundation
The objective is carefully considering how users interact with a system and how the system responds. All the decisions here should be guided by the user’s likely mental model. The goal is to create interactions that are intuitive and predictable.
18.5.2 Understanding User Tasks
Here you define the “what” (what are users trying to accomplish) and the “how”. Here you map out the logical steps taken to reach a goal and identify which tasks are performed most often to prioritize efficiency in the design.
18.5.3 System Support & Interface Metaphors
They bridge the gap between concept and reality/
- System Support
- Provide necessary functionality
- Offering guidance to prevent confusion
- Delivering feedback to confirm successful actions
- Interface Metaphors
- Leverage analogies to the physical world
- Helping users grasp digital concepts through existing real-world knowledge
18.5.4 Interaction Modes and Styles
This includes input methods and interaction styles. You should choose the combination that best supports specific user needs and task efficiency.
- Input Methods (The “How”): Keyboard, Touch, Voice, Stylus
- Interaction Styles (The “Vibe”):
- Direct Manipulation: Dragging and dropping elements
- Conversational: Natural language processing / AI chat
- Menu-based: Structured navigation through lists
18.5.5 The Path to Intuitive Design
- Ensure Predictability
- Focus on the User
- Use Familiar Language
- Provide Feedback
18.6 Activity Based Conceptual Models
Conceptual models can be classified by the primary activities they support.
18.6.1 Giving Instructions
A model where the user actively tells the system what to do by issuing commands. It is used in word processors, CAD systems, vending machines, and consumer electronics.
It is efficient and precise, leading to
- Quick Interaction
- (Efficieny for) Repetitve Tasks
- Predictability
- Minimized Ambiguity
You should use it when the user need to be the active director of the system and for tasks requiring high speed, exact results, and standardized procedures.
18.6.2 Conversing
A model based on the metaphor of human-to-human conservation where users communicate with the system as if it were another person. It moves beyond simple command-response to a dynamic exchange using natural language inputs. GitHub Copilot is an example of this.
It can range from low complexity implementations like menu-driven voice recognition to high complexity implementations like sophisticated NLP dialogue systems. Key example are virtual assistants, search engines, scheduling assistants, and advice systems.
They are good because of their familiarity (everyone knows how to talk and listen) and they are highly effective for novices, technophobes, and universal access (those unfamiliar with technical interfaces). They reduce anxiety and make complex systems feel more human and approachable.
They have disadvantages though:
- Parsing Errors
- Contextual Failures
- The Design Requirement: To be effective, these models require:
Robust Language Processing
Cleat Feedback Mechanism
18.6.4 Exploring and Browsing
A model that supports open-ended exploration of information spaces. It allows for navigation, searching, and discovery in a flexible, non-linear manner. It mimics physical exploration.
Best use cases for it is when the user don’t have a fixed goal or when the objective is to gather general information.
It is used in design tools, mobile tech, spatial data apps, 3D environments, and collaboration tools. Other examples are websites, digital libraries & archives, multimedia systems, and educational resources.
Key characteristics for it include
- Flexible Navigation
- Information-RIch Environment
- Serendipitous Discovery
- Support for Varying Expertise
18.6.5 Comparsion of Activity-Based Models
| Model | User Goal | Primary Benefit |
|---|---|---|
| Giving Instructions | High-precision task execution | Efficiency and Speed |
| Conversing | Information Seeking / Support | Accessibility and Familiarity |
| Manipulating | Spatial Reasoning / Agency | Mastery and Direct Control |
| Exploring/Browsing | Discovery / Information Gathering | Flexibility and Non-Linearlity |
A single system often uses multiple models. The choice of model should always align with the user’s goals and workflows.
- Direct Manipulation: Best for: ‘Doing’ tasks requiring spatial reasoning or iterative adjustments.
- Examples: Drawing tools, flight simulations, or driving interfaces
- Issuing Instructions: Best for: Repetitive, structured, or batch tasks
- Spell-checking, data processing, or configuring system settings
- Conversational Models: Best for: Children, novice/computer-phobic users, or specialized advice systems.
- Phone-based help services or virtual assistants
18.7 Conceptual Models Based on Objects
It is based on grounding digital interactions in familiar physical objects to reduce cognitive load, make complex systems feel approachable, and enable users to learn systems more quickly.
They have multiple advantages including:
- Reduced learning effort
- Predictable interactions
- Enhanced satisfaction
- Hybrid potential: Can be combined with activity based models
18.7.1 The Desktop Metaphor
Organizes digital files, folders, and applications as office materials.
18.7.2 The Spreadsheet Model (Bricklin)
Based on traditional accounting ledgers.
18.7.3 The Star Interface
Organizes applications around a central hub (the “star”).