Conceiving Architectural Design as a User Interface
What if architects focused on the occupant experience to originate their building designs? User Interface (UI) design methodologies, culled from various digital media, may offer architects insight about how to create designs that communicate with occupants directly. By acknowledging the cognitive functioning of the brain, architects can endeavor to utilize geometric configurations to elicit specific feelings and actions.
Put succinctly in Everett N. McKay’s book UI is Communication, “Design is making creative decisions on behalf of target users to create a product that satisfies their goals [p248].” The keywords for architects in this statement is to create spatial environments “on behalf of target users“ – not for themselves, their clients, or on whim.
While planning for the communicative components of project legibility, architects must simultaneously incorporate more traditional requirements. Thus, focusing on the user experience would be a complement to a robust design generation process which may already include architectural ideas from contextual, environmental, cultural, programmatic, code-related, and economic factors.
The Architecture Interface
By defining architecture as an interface between an occupant and the building, architects have more tools to realize the “occupant perspective” as a primary driver of design decisions. This concept of figuratively “communicating with an occupant” is a good technique to elucidate what many architects already intuitively understand and include in their design process.
By substituting the word “building” for UI, we can derive similar insights for architecture from McKay’s text about UIs:
A well-designed [building] boils down to communicating to users in a way that is natural, professional and friendly, easy to understand, and efficient. By contrast, a poorly designed [building] is unnatural, technological and mechanical, and requires users to apply thought, experimentation, memorization, and training to translate it into something meaningful.
From this point of view, [building] design isn’t a subjective visual art about pixels and aesthetics but rather a principled objective communication skill to explain tasks to users. By focusing on communication, design decisions that initially appear subjective – such as control selection, layout, color, and animation – become much more objective. [p3]
Projects which involve user unfamiliarity – those with frequent temporary visitors such as retail, transportation centers, or hotels – have especially important communicative components, typically in reference to circulation and function. For architectural programs that are more static – such as an office space – the spatial environment is also constantly offering hints and cues. Even though an office worker may seem unaffected by their spatial environment seated at a desk for 8 hours a day, the architectural components that immediately surround them still influence their productivity and contentedness. A well-executed contemporary office environment employs a variety of architectural spaces, with elements that encourage collaboration and innovation.
Beaver Workshop Office Space in Beijing by MAT Office, seen in diagram at top of page
“UI form follows communication. . . every visual design element should be justified by what it communicates [p8].” For example, a window may convey a wide variety of instructions when one considers how many parameters an architect will sort through to arrive at the shape, size, transparency, view, and relationship with other architectural components. These factors all communicate an affordance: whether this is a window for looking, a window for illumination, a window to avoid, a window to operate, or a window that acts as an attractor.
Search for Purpose
Through research in evolutionary psychology, it can be deduced that humans are constantly scanning environments for purpose (i.e., “What does this forest afford?” “What does this shelter afford?” “How do I operate this door?”). Users are not often consciously aware they are cognitively processing a particular space to establish legitimate purpose; they will seek cover from the rain or turn a doorknob without conscious thought. But, indeed humans are always scanning to queue their next actions, correlating what their environment affords with what they need.
People are emotional and react emotionally to a product’s visual appearance. Consequently, [a] product should look the part – it should look like it fulfills its purpose well. . . . Don’t assume that users will see the beauty that lies beneath – they won’t. If a UI isn’t comprehensible, little else matters. Confusing is never in style or tasteful. [p5]
Comprehensibility is key to understanding the potential of a user-focused communicative design theory. Humans can interpret a wide variety of visual cues upon which we act socially, verbally, and through movement.
Importantly, design elements should not be placed arbitrarily, but rather should follow the documented condition of how the human eye operates under specific conditions. There are measurable patterns to how humans scan digital interfaces as well as their physical environment.
McKay’s Diagram of UI Scanning, Scanning as an F-Pattern from Tech-Jini
Human Visual fields of Concentration, Eye movement first 2 seconds (Alfred Yarbus, 1967)
So what exactly is happening in an occupant’s brain that we can try to accommodate with design? This is where a lot of the recent cognitive and neurological research aligns with the objectives of UI design. “Users’ expectations about discoverability, affordance, and predictability are set by their prior experiences with all the other software they have used [p27].”
Humans are constantly adding experiences to our individual memories, and when we sense any spatial environment – new or previously encountered – that new space and its components are matched against our prior history with similar environments and components, creating a set of assumptions through which an occupant will act and feel.
Even ostensibly “subjective” design decisions are laced with communicative features. What might seem stylistic or superfluously detailed, is often meant to remind a viewer of some type of indexed cognitive association. A red tile roof, thick mullions, or a green accent wall, all have laced meanings to occupants. A red tile roof can remind someone of a geographic location or climate, thick mullions can signify sturdiness, and a green accent wall can mimic nature. There is always some overlap between aesthetic and purposeful design elements, I write more about it here.
Subconsciously, occupants use their intuition to navigate and prosper in their spatial environments.
Definition: A UI is intuitive when target users understand its behavior and effect without use of reason, memorization, experimentation, assistance, or training. . . For any software project, it’s a sure thing that having an “intuitive UI” is a top goal. For users, describing a UI as intuitive is among the highest praise they can bestow [p21].
An online user may, for example, visit a bank’s website and have a negative impression of their transaction due to confusing buttons or menus. That same customer, when conducting a transaction in the bank’s retail branch, may generate a similar impression of insecurity due to architectural features such as poor lighting, a disorganized layout, or cheap finishes. It should be noted that, in comparison to a digital interface, this customer may struggle to be as consciously aware of their physical environment due to a lesser degree of sophistication.
A UI is intuitive when it has an appropriate combination of discoverability, understandability, affordance, predictability, efficiency, responsive feedback, forgiveness, and explorability. Given that the word intuitive is poorly understood, I recommend against using it in design discussions. . . Instead I recommend using the specific attributes of an intuitive UI when you’re giving feedback. [p26]
For example, one would say that a component or configuration lacks legibility, efficiency, or affordance. This description keeps the focus of designing architectural features on their actionable potential.
Edmunds Headquarters, Santa Monica, California, uses color and glyphs to route occupants
Interface Design Process
In UI, the layout process is typically conducted as separate from and prior to visual element design. Hence in a UI, the intended purpose and expected efficacy of elements is often evaluated by a role one might describe as “interface designers.” It is the inverse in architecture where visual designers lead the design team, perhaps precluding a system to strategize occupant perspectives. Potentially, Architectural design strategies may benefit from including “interface management” in conjunction with the more traditional programming and contextual design exercises.
Use a communication-driven design process to create better designs and make the right decisions quickly and confidently. During the planning phase, develop value propositions, define personas, and determine your top scenarios to understand who your target users are, what they are likely doing, and why they care. [p9]
Typically, architecture professionals and students tend to treat humans abstractly, goading generic ideas of humans into glorified spaces and enamoring them with decorative features. This type of process does not identify an occupant’s particular needs and aspirations, how those needs align with a client’s needs, or how goals might change over time.
IntuArch’s Persona/Scenario/Storyboard Process
One way to incorporate interface design into the architectural design process might be to create a mock system of Personas, Scenarios, and Storyboards. At the base level, Personas are stereotyped humans who represent particular demographics or actions. “A Persona is a fake person constructed using real user research data to represent a class of real target users [p255].” Personas can also have deeper descriptions of the types of cognitive memories being brought into the space under consideration, including their expected previous history with similar buildings and their history with the particular context under consideration.
An interface manager would then place these Personas at different junctures within a design project to see how architectural features influence their behaviors and impressions. “A Scenario describes a specific [Persona] trying to achieve a specific goal or perform a specific task in a specific context [p253].” A project as seemingly straightforward as an apartment complex has several Personas who encounter the spatial environments entirely differently. Besides the residents, apartment building Personas include prospective tenants, guests, staff, and maintenance. The residents themselves can be further divided into separate Personas to account for demographic differences and cognitive associations. Each Persona’s experience can contribute significantly to the project’s financial success.
Hangzhou NO.2 School of Future Sci-Tech City in Hangzhou, China by LYCS Architecture
These Scenarios can then be placed sequentially into Storyboards. The Storyboards can aid or amplify other architectural diagrams, such as those for program, context, or circulation. (I wrote an article about how to create effective diagrams here.) Interestingly, McKay utilizes the term “wireframe” to describe the very similar tool of “architectural diagram,” creating a framework system to relate ideas to visual elements.
At the wireframe level, every visual design element should be justified by what it communicates. A wireframe is a prototype with content and functionality represented by rough controls, rectangles, lines, text, and simple glyphs. Layout and element sizes are rough, and details such as colors, backgrounds, fonts, graphics, and icons are often omitted. [p129]
The Tinke Mobile Fitness App Wireframe
Importantly, the Persona/Scenario/Storyboard system needs to be considered before concept designs are imagined in order to provide an effective means for design evaluation. “You can’t use scenarios to determine the right solution if the solution is already baked into the scenario. Good scenarios focus on users’ goals, their problems, and their context, without providing any solutions [p254].”
Other Testing Strategies
McKay also arrives at a few other helpful strategies to ensure final designs relate well to human experience.
Users know their goals but not how to achieve them. They know the task destination but not the starting point or the steps to get there. Assume that they know only the most essential data from memory and need help with everything else. Without prior knowledge or experience, users don’t know what your program does, what tasks it performs, how it works, or that they can trust it. [p56]
Take a view, a series of views (Storyboard), or a planar drawing, and circle every element to evaluate them for visual information to ensure purposeful communication. The process could be similar to “redlining” in the design phase, but instead of highlighting changes would identify or edit interpretable architectural features.
Sketches and Notes for a UI Wireframe
“A five-second test evaluates the scannability and memorability of [an architectural space] simply by having users view it for five seconds and then answer a few basic questions about the purpose of the [architectural space] and important design details, such as the call to action [p283].” In architecture, I have a similar evaluation I call the “squint test.” If you close your eyes or blur your vision, can you still reference the major geometries of both the space you are in, and all spaces throughout your journey, that are integral to your navigation and action?
How Users Figure Things Out
- Users will assume that your program has design patterns and interactions that are standard for its environment.
- If users are familiar with a similar [building] from another feature or other program, users will assume that the current feature has similar interaction.
- Unless they perform a task frequently, users won’t remember exactly how to perform it from memory and will have to relearn it each time.
- If an action is discoverable, looks relevant, and has a clear affordance, users will try it right away.
- If an action doesn’t have an affordance, users will try it only after eliminating all obvious alternatives first.
- If an action isn’t discoverable, novice users won’t find it.
- Users don’t read, they scan, as Steve Krug points out in Don’t Make Me Think!. That is they don’t read [buildings] completely for comprehension; rather, they scan to find what they are looking for quickly.
- Users need to gain confidence as they perform a task and are reluctant to proceed with a task or make commitments without building sufficient confidence. Users need to know where they are at each step in a task. They need clues to confirm that they are in the right place. [p58-59]
- Users assume that attractive products are better designed and more usable (this is known as the aesthetic-usability effect). Don’t assume that users will see the beauty that lies beneath; they won’t. [page 131]
McKay devotes a substantial amount of space within his book to cataloging the common types of controls – such as sliders, textboxes, radio buttons, checkboxes, links, etc. – the options for their graphic expression, and how users might infer different outcomes accordingly. Might the individual components of architecture offer a similar opportunity for categorization and communication possibilities? In addition to the previously described example of the parameters of a window, the stairs, elevators, atriums, arches, floor heights, soffits, rails, doors, skylights, etc., all have different affordances, messages, and associations. It could be insightful to map each architectural component for communicative qualities.
A “Flat Kit” for UI Design Layouts
In a communicative design, architectural components could be reclassified by their performance with tasks and navigation. Performative categories might include landmarks, pathways, nodes, volumes, areas, and edges. These elements are aided by standard visual factors, such as bold colors and recognizable shapes, and can be arranged by employing standard architectural patterning tools such as simplification, hierarchy, sequence, contextual association, phenomenal transparency, grouping, and scale. Together these elements and their details are the entirety of an occupant’s cognitive mapping.
The following sections discuss the need for this type of analysis in architecture, and how it relates to several ongoing cultural paradigms.
Why Forecasting Users is Important
There are many anti-human movements encroaching into contemporary architectural practice, none of which treat occupants with empathy.
- Codes. The increasing complexity of codes has made the design of buildings a calculations-oriented process, prefacing energy, zoning, and egress standards.
- BIM. The BIM process prioritizes the descriptive nature of the architectural components – the materials and manufacturers – and their tectonic relationships.
- Heroes. Architectural education insists on uniqueness in design, promoting the “hero” architect as an aspirational goal.
- Parametrics. Breakthroughs in academia have been concentrated in the potential of parametric modelling, which, similar to BIM, is primarily about the relationship between architectural components (though it may sometimes involve environmental or user data).
- Renderings. The prioritization during the design process to procure pretty renderings, which evaluates architecture from only one static point of view, disabling an experiential analysis. Or, think of the latest developer marketing video you might have seen. Did that “human” take a natural circulation path? Probably not, their route is structured so that you see fancy features.
If your goal is to improve the design, the last thing you want is the presentation of your prototype to discourage feedback. I find that people react emotionally to beautiful things and that beauty hides flaws. Consequently, beautifully rendered prototypes often receive less critical feedback than rough ones – which isn’t what you want here. [p271]
It would be key to develop presentation material to solicit feedback, not to prove beauty. I am already a proponent of creating diagrammatic representations of a project’s major design ideas for relaying information to clients; this viewpoint from McKay’s book offers a clarification of why that’s so important.
It’s not all bad news though. Students and professors at many institutions have been testing design options and studying user decision making through virtual reality environments, often in relationship with real world spatial counterparts. Though there are concerns about the accuracy of virtual reality experiences in relationship to real ones, there is tremendous potential to provide user feedback to architects during the design process.
No Utopia. As with any architectural theory, designing architecture as a UI is not a utopian or eternal solution. Ideal UIs will change over time because culture and behavior change over time. It is of course much easier to update elements in a digital interface than it would be in a completed building. Ideally, interface management would contribute responsive design elements that prove resilient with time.
Feedback. Relative to measuring performance within a UI, there is less feedback for poor architectural performance. Getting lost is perhaps an obvious feedback for an occupant to relay to a designer. But, it is much harder to measure comfort, productivity, interaction, purchasing, etc. Though it has always been difficult to measure the value of a particular design in reference to other design options, architects and clients should always remember that there is an economic cost to confusing or abrasive buildings.
Signage. UI employs a lot of text, which makes its design strategies literally communicative in the first place. Even so, contemporary digital designers have moved towards icons and graphics to guide user decision making.
Not every UI needs a main instruction, and many don’t have them. The key: If a user were to ask you what to do in person, would you bother to say the instruction? If not, the page is better off without it. [p50]
Written communication – ie, signage – in architecture is of course often warranted and necessary. There is no doubt that, for example, airports require signage to route passengers. But for most programs, architects can challenge themselves to use geometric forms that encourage or respond to occupant behaviors.
Digitization of Our Physical World?
When digital platforms were first established, designers employed skeuomorphic graphic representations of real physical features for humans to cognitively interpret the functionality of various actions. A prominent example might have been the iPhone’s app bookshelf. Skeuomorphic representation has since waned considerably, as most current designs are relatively “flat.”
Apple’s former homepage “Bookshelf”
But, could design inference work in reverse? As we spend more time on screens, might we expect our physical environment to mimic the digital? Millennials may already be treating digital interfaces as their primary means of communication with other humans. But, they are also using digital media as their primary means of communication with non-living entities such as products and brands. If augmented and virtual reality become commonplace in the near future, it would be wise for architects to study these cognitive relationships. If users are, for example, conditioned to scan a digital interface for actionable information, might architects have some obligation to provide more inferrable information in architecture as well?