A Top Building Researcher Asks: Why is Architecture Afraid of Science?
Recently we’ve written a fair amount about the state of architectural research. The general consensus appears to be that it lacks rigor and, even more importantly, is not grounded in good science. Steven J. Orfield has some strong opinions about architectural research. He’s been conducting it—for architecture and design firms, as well as Fortune 500 companies—at his Minneapolis-based, Orfield Laboratories, for more than three decades now. Late last week I talked to him about why architects are afraid of science, how he would introduce it into the schools, and his work in the field of universal design.
MCP: Martin C. Pedersen
SJO: Steven J. Orfield
Your lab has worked with a range of clients, both corporate and architectural. Why do other industries use and rely on science and research to a much greater extent than architecture?
I think almost all industries have a greater engagement in research than architecture does. Why? Because architecture really doesn’t do research. The field produces many PhDs. It provides Master’s degrees and a lot of graduate education, but most of it is theoretical and historical. Architecture doesn’t teach science, because it doesn’t know science. And architecture has an inherent belief—at least according many of the architectural deans that I know—that undergraduate American students don’t want to learn research, they don’t like science, and all they want to do is “art.”
For the 46 years that we’ve been in business, we’ve hosted tours from architecture and design schools around country, and we know that to be false. The students love science. They just don’t get any good science in school. The problem is that science and design don’t have a common language. If you’re going to teach science to architectural students, you need to use experiential methods. That requires you know all of the science as it relates to human perception and comfort. And that you can create scientifically-grounded demonstrations, because you understand how to do it. To use science effectively, you have to first understand it.
If an architecture dean came to you and said: create a research program for students, what would that look like?
It would be based on the same concepts that we use with our clients. We bring clients in and spend four to five hours bringing them through an immersion session. With graduate students, we would do the same immersion over the the period of the course. We’d develop a curriculum that moves from perceptual demonstrations related to the design process, and then once the students understood those intuitively, we would teach them aspects of human perception and cognition. Like clients, students need an experiential base where they can teach themselves by observation, and then learn the formal practice.
Why is architecture so reluctant to embrace research and science?
It’s foreign to them. They’re also afraid that if they brought research into the “design” process, they would get marginalized. Architects have never found it necessary to learn to design buildings inspired or informed by science, because they believe that their intuitive process works well. And they have no useful way to compare this to the more scientific research-based design method, as they have no experience with it.
Except for the occasional energy performance study, why is there so little post-occupancy research done on buildings?
The really interesting thing is, there is some post-occupancy research done, but hardly any pre-occupancy research. And if you do post-occs but haven’t done pre-occs, you’ve got nothing for comparison and for judgements.
Which is essentially the basis of the scientific method?
Right. We’ve done pre-occs and post-occs for probably twenty-five years, and we developed a model for work spaces that measured the perceived organizational quality, job quality, compensation quality, and general and local environmental quality. What our studies measure is how you feel about your company. And what aspects of how you feel about your company relate to the building. It’s the only study type we know of that actually measures work satisfaction instead of workplace satisfaction.
So you’re not measuring the architecture, per se?
We’re measuring more than that. We tell our clients: If your organizational quality is perceived as low, there’s really no point in building a new building. If your compensation quality is perceived as low, there’s no point in building a building. Most post-occupancy studies are to a large degree in place to support the building of buildings. Our studies are in place to support organizations and how they think about their employees’ health and well-being.
Does your lab do energy modeling?
We do energy modelling on lighting and daylighting. But the energy piece is secondary to the perceptional qualities of lighting and daylighting. Our argument is that there are two main things you want out of a building: you want to be perceptually comfortable, and you want the building to be emotionally resonant. You want to like the place and be comfortable in it. There are different methods to get to each of those goals. Perceptual comfort can be derived from good scientific research, just as our building performance standards are (acoustics, AV, lighting, daylighting, thermal comfort, indoor air quality). Perceptual preference comes from quantitative subjective research on the emotional and associative response to the building design. This is accomplished by POP (Perceptual Occupant Programing) visual juries which expose occupants to images of the schematic design of the building interior and exterior, while they use semantic differential ranking scales to rank a set of bipolar attributes, such as “comfortable – not comfortable” or, “low stress – high stress.” During these POP juries, there is no discussion or ranking of opinions, as in focus groups, or what architecture calls, “design charrettes.”
Who are your clients? Is it architectural firms, or building owners?
About half the time we’re hired by owners, and the other half by designers. The owner-client tends to be open to a much broader involvement. The architectural scope is controlled by the architect’s often narrower interests. When I’m hired by an owner to do a new office building, I can do a whole range of things: architect selection competitions, pre- and post-occupancy studies, acoustics, lighting, daylighting, thermal comfort, indoor air quality and performance commissioning. If I’m hired by the architect to do the same project, he or she may be interested in the acoustics and the lighting, or the audio-visual system design.
Why the difference in research scope?
Designers tend to believe something is important if the client says it’s important. An architect might ask the client, “Are acoustics important to you in this project?” If they respond, “Not particularly,” then the architect checks that off their list. Rather than teaching the client what they need, architects generally ask clients what they want. As an expert testifying to design failures in court, I’ve heard this in testimony a number of times.
It’s almost like they get the research that they’re asking for, which is tight and limited.
Clients don’t know what to ask for. It’s like going to the doctor, and the doctor asking, “Is there anything about your health that you’re interested in?” We always assume that the client has no idea what they need. And we’re not governed by their initial impressions of what they want. Our job is to do for that client what they would do for themselves, if they had our level of information about design research and testing.
What’s an area of research that you’d like to explore?
There’s very few areas that we don’t have our hands in, but I love psychology. And we’ve worked with many psychologists around the country, cognitive psychologists, perceptual psychologists and others. We’d like to get into more of those collaborations. I’m interested in neuroscience. We study that at some depth. But we don’t think the work coming out in that field is very high resolution, as yet.
It’s an emerging field.
And because of that it has limitations. If you’re trying to design for a particular group, for example, neuroscience doesn’t take into account occupant demographics, to a strong degree. Neuroscience is an extremely limited, low resolution tool, and architecture is trying to use it for an extremely high-resolution set of problems. If we do POP visual juries, in architecture, we can get into the same things that neuroscience is after, much more easily, with higher resolution, better demographic control, and less expense than they can. At some point, these tools should merge and appear in design offices as subjective testing systems, but this would require architects or technicians who understood the subtlety of research.
But you will concede that neuroscience is an emerging field that holds promise for architecture?
The problem with neuroscience is, it doesn’t necessarily tell you much. A certain area of your brain lights up. What does that tell you? Did that stimulate you? Did you like or dislike something? Neuroscience is trying to map those things out, and eventually it will be an interesting tool. But it will become an interesting tool when a client can walk into architect’s office, put a headset on, look at a set of images, and go through a process that predictively gives the architect an idea what the client wants, in terms of the building’s resonance.
Neuroscience may eventually be good for producing better aesthetic judgements by architects, meaning judgments that are more sympathetic to the user. On the other hand, neuroscience won’t do anything about perceptual comfort: acoustics, daylighting, thermal comfort, indoor air quality, the things everybody complains about in buildings. Those are issues where we already know the science. We know what the goals should be. We know how to design to them. But more than 99% of projects in the United States don’t use building performance science.
You’re doing a lot of work on the universal design front. Tell us about that.
We’ve spent a large portion of the last ten years studying cognitive and perceptual disabilities, and developing standards for building performance, related to those disabilities. So, we’re working toward developing a universal design standard for perceptual and cognitive disabilities that we know will be much better for those with disabilities, but it will also be preferred by those without. What we’re saying is, buildings should be designed with certain kinds of limits, related to perceptual stimulation and cognitive complexity. Those should be rules for all buildings. Because they’re better for people that way. And since the science is fairly clear that they’re better for people that way, it’s no longer a question of what “style” of building you want, or what the architect is trying to evoke in his or her design. What we want to know is—and the only thing we want to know is—how do we make that building wonderful for the user?
And that’s style agnostic?
It’s not style agnostic, at all. But the styles that are applied, the aesthetics, have to be resonant to the user, not just the architect. I believe strongly in aesthetics, but I also believe—and this is something that many architects really dislike—that you can measure aesthetics. We’ve done it over and over again. We believe that what the architects think they’re doing, as art, can be measured in the user to determine whether it works. And often it doesn’t.
We’re starting to believe that buildings all over the world should be designed with more of a peaceful aesthetic, with low perceptual noise levels and low cognitive complexity levels. Buildings that are simple, that are easy to make your way through, they won’t have a lot of visual or acoustic noise. Buildings that provide a gentle experience and keep your threshold of stress down. That’s rarely done intentionally.
It’s a great irony that most of the big companies who design products have user-experience departments. But architecture, perhaps the most complicated science-art of all, doesn’t have a user-experience department. It’s often practiced as poor quality art, with no user measurements. We’ve done user experience work for Whirlpool, Harley Davidson, Black and Decker, Microsoft, and we find that the concept of user-experience science is completely foreign to the design community. We also find, somewhat surprisingly, that it’s foreign to much of the product community as well. The industrial design field is like the architectural world. It absolutely hates being measured from the outside.
Featured image of Orfield, inside the anechoic chamber, at the Orfield Laboratories, via WCCO.