Capitolo 4
Conceptual Models: The Mental Maps We Create
We navigate the world by creating mental models of how things work - these are the internal representations we build to understand and predict the behavior of systems and objects around us. Good conceptual models allow us to predict the effects of our actions with reasonable accuracy, even when encountering something new. Consider a strange tandem bicycle with both riders facing forward but with the second set of pedals connected to the front wheel-you immediately know it won't work because you can mentally simulate its operation and recognize the steering conflicts that would arise. This immediate recognition demonstrates how we use existing mental models to evaluate new situations.
Scissors work intuitively because their operating parts are visible and their affordances (finger holes), constraints (hole sizes), and mappings (relationship between handles and blades) are clear. The direct mechanical connection between handles and blades creates an immediate understanding of cause and effect. In contrast, digital watches with unlabeled buttons provide no such clarity-there's no visible relationship between controls and functions. Users must memorize arbitrary button sequences, leading to frustration and errors. This contrast highlights how visible mechanisms support natural learning while hidden ones require explicit instruction.
The refrigerator example perfectly demonstrates how manufacturers can provide false conceptual models. With two compartments and two labeled controls (one for the freezer, one for the refrigerator), users naturally assume each control independently adjusts its named compartment. In reality, one control adjusts the thermostat while the other adjusts air distribution between compartments, making temperature adjustment frustratingly difficult. This mismatch between the user's mental model and the actual system operation leads to persistent confusion and inefficient use. Even experienced users might never fully understand the true relationship between the controls and their effects.
People form mental models through experience, instruction, and interaction, interpreting a device's visible structure (what Norman calls the "system image"). When this image is incoherent or inappropriate, the device becomes difficult to use. This explains why people develop "folk theories" about how things work-like believing that turning a thermostat to maximum will heat a room faster, when in reality thermostats are simply on-off switches that activate heating systems at full power until the set temperature is reached. Other common folk theories include repeatedly pressing elevator buttons to make them arrive faster or holding down phone buttons harder to get better reception.
The key insight isn't that people have wrong theories, but that everyone creates mental models to explain their observations, especially when designs provide no hints about correct operation. Good design creates accurate mental models by making the system's actual operation visible and understandable. This can be achieved through various means: clear visual feedback, logical relationships between controls and their effects, and consistent behavior patterns. For example, modern car dashboard displays often show real-time feedback about fuel efficiency, helping drivers develop accurate models of how their driving behavior affects fuel consumption. Similarly, well-designed smartphone interfaces use animations and visual cues to help users understand where their content "goes" when they perform actions like deleting or archiving.
Capitolo 5
The Seven Stages of Action: How We Interact With Objects
To understand what makes tasks difficult, Norman breaks down human action into a seven-stage process that illuminates how people interact with objects and systems in their environment. The basic structure involves forming a goal, taking action in the world, and checking results. Goals must be transformed into specific intentions, which become action sequences that control our muscles through a complex cognitive and physical process.
The complete seven stages are: forming the goal (deciding what we want to achieve), forming the intention (determining what actions will help achieve the goal), specifying an action (planning the detailed physical movements), executing the action (performing the physical movements), perceiving the world's state (using our senses to observe changes), interpreting that state (making sense of what we perceive), and evaluating the outcome (comparing results with our original goal). This model reveals two critical gaps in human-object interaction that designers must address.
The Gulf of Execution represents the gap between a person's intentions and the actions allowed by a system. This gulf is measured by how directly the system permits intended actions without extra effort or confusion. For example, early microwave ovens required users to understand power levels, cooking times, and multiple button sequences - creating a wide gulf of execution. Modern microwaves bridge this gulf with preset buttons for common foods and smart sensors that automatically adjust cooking parameters. Similarly, car interfaces evolved from complex manual chokes and timing adjustments to simple push-button starts that handle engine management automatically.
The Gulf of Evaluation measures the effort required to interpret a system's state and determine if expectations were met. This gulf is small when a system provides easily accessible, interpretable information that matches the user's mental model. Digital thermostats illustrate this concept well - older models showed only current temperature with cryptic programming interfaces, while modern smart thermostats display clear visual feedback about temperature trends, scheduling, and energy usage. Norman notes that while VCRs improved on film projectors by making incorrect cartridge insertion obvious, they weren't perfect. He shares an anecdote about conference technicians troubleshooting a VCR for several minutes before discovering there was no tape inside-a design flaw that provided no visible indication of whether a tape was loaded once the door was closed.
These gulfs explain why people struggle with everything from water faucets to digital watches to smartphone interfaces. Good design bridges these gulfs through four key principles: visibility (making important information and controls obvious), good conceptual models (helping users understand how the system works), clear mappings (making relationships between controls and their effects obvious), and continuous feedback about action results (providing immediate, clear information about what the system is doing). Modern touchscreen interfaces exemplify these principles by showing direct manipulation of on-screen elements with immediate visual feedback.
Capitolo 6
Knowledge in the Head vs. Knowledge in the World
Knowledge doesn't need to be stored entirely in our heads-it can be distributed between mind and environment. Norman illustrates this with multiple examples, including the classic anecdote about a car requiring an unintuitive action (being in reverse) to remove the ignition key. This demonstrates how some knowledge must reside in memory rather than being discoverable through interaction. Another example is the QWERTY keyboard layout - while not intuitive, once learned, it becomes internalized knowledge that enables efficient typing.
We navigate this tradeoff constantly, deciding what to memorize versus what to reference externally. This explains why people function well yet struggle to describe their actions-like navigating familiar routes without being able to verbalize directions. A skilled driver smoothly operates a manual transmission without conscious thought but may have difficulty explaining the precise timing of clutch engagement. Norman distinguishes between declarative knowledge (facts and rules, easily taught) and procedural knowledge (skills like playing music, best learned through practice). This distinction is crucial in understanding how people learn and interact with devices.
The distinction between knowledge in the world and in the head offers important design insights. External knowledge works best when relationships between information and actions are natural and easily interpreted, such as push bars on doors that indicate where to apply force. However, internalized knowledge enables faster, more efficient performance, as seen in expert users who can operate complex software using keyboard shortcuts rather than searching through menus. Good design balances these approaches, letting users rely on whichever is more accessible without impeding experienced users who have internalized operations.
This balance explains why natural mappings are so powerful-they dramatically reduce memory requirements by leveraging spatial relationships between controls and their functions. Norman uses stove burners and controls as a prime example: with arbitrary arrangements, users must memorize 24 possible control-to-burner relationships; with partial mapping (left controls for left burners, right for right), this reduces to 4 possibilities; but with full natural mapping (controls arranged in the same pattern as burners), no memorization is needed at all. Similar principles apply to car dashboard controls, elevator buttons, and light switch panels.
Understanding this distribution between internal and external knowledge allows people to function effectively despite limited memory capacity. It explains why we can operate complex devices without understanding their internal workings, and why we often rely on environmental cues rather than memory. This also explains why standardization becomes so important when natural mappings aren't possible-standardized mechanisms need only be learned once. Examples include traffic signs, emergency exit symbols, and computer interface conventions that, once learned, transfer across different contexts and cultures. The power of standardization is evident in how international symbols can effectively communicate across language barriers, reducing the cognitive load on users worldwide.
Capitolo 7
To Err Is Human: Designing for Imperfection
People make errors routinely-stumbling over words, repeating phrases, or stopping mid-sentence-yet human language has built-in correction mechanisms that make these errors nearly invisible. Artificial devices lack this tolerance; one wrong button can cause chaos.
Errors fall into two fundamental categories: slips and mistakes. Slips occur during automatic behavior when subconscious actions get derailed, while mistakes result from conscious deliberations gone wrong. The distinction is clear when analyzing the seven stages of action: form the right goal but mess up the execution, and you've made a slip; form the wrong goal, and you've made a mistake.
Most everyday errors are slips-intending one action but doing another, like unbuckling your wristwatch instead of your seatbelt. These "psychopathologies of everyday life" typically occur during skilled behaviors rather than during learning. Slips often result from divided attention; people can consciously attend to only one primary thing at a time, yet we frequently multitask.
Norman identifies six types of slips: capture errors (when a familiar activity takes over an intended action), description errors (performing the right action on the wrong object), data-driven errors (when environmental stimuli override intended actions), associative activation errors (when internal thoughts trigger unintended actions), loss-of-activation errors (forgetting what you're doing midway through an action), and mode errors (when devices have different operational modes where the same control performs different functions).
Good design can both prevent slips and help detect and correct them. Computer systems often request confirmation before destructive actions, but these prompts come immediately after users have committed to the action, making them ineffective safeguards. A better approach is to eliminate irreversible actions entirely-like moving deleted files to a temporary holding place for later recovery.
The designer's philosophy should approach errors not as failures but as approximations of correct action. Systems should be designed to accommodate human fallibility through forcing functions (physical constraints that prevent errors), making actions reversible, and providing clear feedback about system state.
Capitolo 8
The Design Challenge: Balancing Competing Demands
Designers face multiple competing demands that create complex trade-offs. Manufacturers demand economical production methods and minimal material costs, while retailers require products that stand out on shelves and maintain profit margins. Purchasers exhibit different priorities at different stages - emphasizing aesthetics and price during purchase but functionality and reliability during use. Meanwhile, repair services need accessible components and standardized parts. These stakeholders often pull designs in opposing directions, yet successful design must somehow satisfy all these requirements simultaneously.
The complexity of the design process extends far beyond initial appearances. Consider the humble water faucet: its design must address water conservation, ease of use, hygiene, durability, and cost-effectiveness. Public faucets exemplify these challenges - the need to prevent waste led to spring-operated mechanisms that automatically shut off, but these created new problems like insufficient time for thorough hand washing. Sensor-based solutions emerged to address these issues but introduced new complications: unreliable detection, battery maintenance, and the confusion of invisible controls. Temperature control becomes particularly problematic with automatic systems, as users lose direct mechanical feedback.
The myth of the "average person" presents another fundamental design challenge. While physical anthropometry provides detailed data about human dimensions, designing for statistical averages effectively excludes significant portions of the population. Even designing for the 95th percentile still leaves out millions of users worldwide. This has led to the rise of adjustable design solutions - from office chairs with multiple adjustment points to modular equipment that can be customized for different body types and abilities. However, adjustability itself introduces complexity and potential failure points that must be carefully managed.
Cognitive limitations further complicate design challenges. Human attention is a finite resource - when focused intensely on one task, people become remarkably blind to other stimuli, even obvious ones. This "selective attention" phenomenon has serious implications for safety-critical designs. Aircraft cockpits, medical equipment, and industrial machinery must be designed with this limitation in mind, incorporating redundant safety systems and forcing functions that prevent dangerous operations when attention is diverted. Visual and auditory alerts must be carefully calibrated to break through attention tunnels without causing alarm fatigue.
Technology's evolution follows a predictable but challenging pattern: the U-shaped complexity curve. New technologies typically start complex, become simplified as they mature, but then grow increasingly complicated as manufacturers add features to differentiate their products. This pattern is evident in everything from automobiles to smartphones. While modern devices offer unprecedented capabilities, they often become too complex for efficient use. A contemporary car dashboard may offer hundreds of features but can overwhelm drivers with options. The paradox of technology is that attempts to simplify life through added functionality often result in increased cognitive load and learning requirements. Successful design must navigate this paradox by carefully managing complexity and ensuring that additional features truly enhance rather than diminish the user experience.
Capitolo 9
User-Centered Design: A New Philosophy
User-centered design focuses on making products usable and understandable by prioritizing users' needs. Good design should make actions obvious, keep system states visible, provide clear feedback, and maintain natural mappings between intentions and actions. The goal is ensuring users can figure out what to do and understand what's happening.
Norman outlines seven principles for transforming difficult tasks into simple ones:
1. Use both knowledge in the world and in the head, balancing external information with what users have memorized.
2. Simplify task structures by providing mental aids, making invisible processes visible, automating while preserving the task's essence, or fundamentally changing the task's nature.
3. Make things visible to bridge execution and evaluation gulfs, ensuring actions and their effects are perceptible.
4. Get mappings right by exploiting natural relationships between controls and their functions.
5. Exploit natural and artificial constraints to guide users toward correct actions.
6. Design for error by assuming any possible mistake will be made and planning accordingly.
7. When all else fails, standardize, creating consistent patterns that need only be learned once.
Sometimes good design means deliberately making things difficult, as with child-proof medicine bottles or security systems. The principles of good design can be systematically reversed to create appropriate difficulty: hiding critical components, using unnatural mappings, requiring precise physical manipulation, withholding feedback, and making system states difficult to interpret.
Tools profoundly shape not just how we do things but our view of ourselves, society, and the world. The evolution of writing tools demonstrates this-from quill and ink that produced careful, embellished sentences to keyboards that nearly match thought speed, each technological advancement subtly altered written expression itself.
The ultimate goal of design isn't just to make things easier-it's to improve quality of life by creating objects that work with human psychology rather than against it. Good design makes operations obvious, exploits natural constraints to guide users toward the right action, and acknowledges that errors are inevitable. When we encounter well-designed objects, they feel like natural extensions of ourselves, enabling us to focus on our goals rather than struggling with the tools meant to help us achieve them.
