For decades, usability has been one of the central pillars of Human-Computer Interaction. Whole industries formed around improving it, organizations hired usability specialists, and universities built research labs to study it. The effort produced real progress. Interfaces became cleaner, navigation improved, and tasks that once required training manuals became manageable through visual cues.
Yet a persistent tension remained. People often reported that systems were easy to use while expressing uncertainty about how those systems actually worked. They could follow the steps and complete the task, but the logic underneath stayed unclear. The tension reveals an important distinction: ease of use and conceptual understanding are not the same thing. A system may be operationally simple and still leave people conceptually disoriented, and that gap is where cognitive distance becomes visible.
4.1 The Rise of Usability as a Discipline
Early usability research focused on efficiency
Usability research grew from early studies of how people interact with machines. As computers spread through workplaces in the late twentieth century, researchers began measuring how quickly people could complete tasks with different interface designs (Card, Moran, and Newell 1983). The measures were quantitative: task completion time, error rates, and the number of steps needed to finish a task. They exposed inefficient structures and showed that careful design could substantially reduce the effort of using a system.
Jakob Nielsen did more than anyone to bring these ideas into mainstream practice. Through books, research and consulting, he popularized evaluation methods designers could apply quickly, and showed that even simple testing with a handful of participants could uncover major design problems (Nielsen 1993).
The emergence of usability heuristics
To help designers find common problems before testing, Nielsen and Rolf Molich proposed a short set of usability heuristics, later refined by Nielsen into the list most practitioners know today (Nielsen and Molich 1990; Nielsen 1994). They include visibility of system status, a match between the system and the real world, consistency and standards, and error prevention. Each names a common source of friction, and when systems violate them people tend to become confused or frustrated. The heuristics remain highly influential because they give teams a fast, practical way to evaluate an interface before it reaches the public.
Usability success stories
Many widely used products improved dramatically through this work. Early computer interfaces required people to understand file paths and command syntax; modern ones hide those details behind visual metaphors and simplified navigation. The graphical interface popularized by the Apple Macintosh, building on the Xerox Star, made windows, folders and icons the everyday vocabulary of computing, letting people work with familiar conceptual structures instead of abstract commands (Smith et al. 1982).
These successes demonstrated the value of human-centered design. They also reinforced a particular way of thinking about usability, one that associated it above all with efficiency and simplicity.
4.2 When Usability Metrics Hide Deeper Problems
Task completion does not guarantee understanding
Usability studies typically rely on task-based evaluation. Participants perform specific actions while researchers record success rates and completion times, and if most participants succeed the interface is considered usable (Hornbæk 2006). This method does not always reveal how well people understand the system. A participant may succeed simply by following visible cues: clicking the only obvious button, or tracing a highlighted path through the screens. By the metrics, the interaction succeeded. But the participant may still lack a clear model of how the system works, and if the interface changes slightly, or the task appears in a different context, they may struggle again. The evaluation captured surface performance, not conceptual understanding.
The phenomenon of procedural learning
Many digital interactions rely on procedural learning rather than conceptual understanding. People memorize a sequence of actions that reliably produces the result they want. An employee using enterprise software may remember that generating a report means opening a particular menu, selecting two options and clicking a confirmation button. They know the sequence works, but not why the system is structured that way, and if the menu moves or the interface is redesigned the procedure breaks and they must rebuild it from scratch. Cognitive psychology has documented this dissociation for decades: people can improve at a task without acquiring verbalizable knowledge of the rules that govern it (Berry and Broadbent 1984). The system never became part of a coherent mental model. It remained a set of instructions.
Usability optimization can simplify the wrong structure
Teams sometimes try to reduce cognitive load by simplifying interface elements without reconsidering the structure underneath. Buttons get larger, menus get shorter and visual clutter disappears. These changes can improve efficiency, but if the conceptual structure is still misaligned with how people reason, the fundamental problem persists. The interface becomes easier to operate while still requiring people to adapt to its logic. The experience feels smoother without becoming more understandable, and the cognitive distance is largely unchanged.
4.3 The Limits of Simplicity
Simplicity can conceal complexity
Modern interface design celebrates minimalism. Clean layouts, limited options and streamlined workflows are widely treated as hallmarks of good design, and they can indeed reduce cognitive load. But minimalism sometimes conceals complexity rather than resolving it. When systems hide too much, people lose the ability to see how actions connect to outcomes. The interface looks simple while remaining conceptually opaque, and people can perform tasks without being able to explain the logic behind them.
Automation reduces visible effort but increases abstraction
Automation complicates the relationship further. Automated systems remove steps people used to perform themselves, such as organizing files, recommending actions or generating content, and this often improves efficiency. But it also introduces new conceptual layers. When a system acts on the user’s behalf, the user may lose sight of how the action happened. The interaction becomes effortless but less transparent. Generative AI tools that produce complex outputs from a single prompt are the clearest current example, and Chapter 9 examines them in detail. Cognitive distance grows when system logic becomes invisible.
The illusion of intuitive design
Designers often describe successful interfaces as intuitive, implying that people can operate them naturally without conscious effort. But intuition does not appear spontaneously. It emerges from familiarity: an interface feels intuitive when it matches patterns people have already learned elsewhere. When designers assume their own sense of simplicity is universal, problems follow, because what feels obvious to one group can feel unfamiliar to another. Without research into users’ actual mental models, “intuitive” becomes a misleading word.
4.4 A New Question for Usability
None of this suggests abandoning usability principles. Efficiency, clarity and error prevention remain essential. They are simply not sufficient on their own. Usability traditionally asks a straightforward question: can users complete the task? Cognitive distance adds a second: do users understand what the system did, and why?
The second question shifts the focus of design. It asks researchers and practitioners to examine not only how easily people perform actions, but how well those actions align with their reasoning. When usability and conceptual alignment move together, interaction is both efficient and meaningful. When they diverge, systems can look successful while quietly producing confusion.