Autumn 2005

Making visible the invisible

Can designers and scientists teach each other how to express new concepts in text and image?

Scientists have used images as tools for analysis and discovery for centuries. Compared to today’s scientists, however, Leonardo da Vinci, Andreas Vesalius and Robert Hooke had it easy. The discoveries that scientists are making today are no longer synonymous with what they can see, or even sense. Much of what remains to be discovered about the physical world is hardly palpable, but is speculative, elusive, liminal or invisible. Teams of technicians must be trained to generate and process the occult data that comes from radio telescopes, electron microscopes, magnetic resonance imagers, sonograms, gamma ray spectroscopes and countless other imaging technologies. Advanced tools such as these now allow researchers to learn more about the ocean floor, the inside of a human body and the behaviour of sub-atomic particles, yet they also demand unprecedented skills of interpretation and visualisation.

‘All images derived from data beyond the spectrum of visible light are technological illusions, tweaked conspiracies between human and instrument,’ warns Douglas Smith, an exhibition developer at Boston’s Museum of Science. ‘Viewer beware.’

For research scientists, the pressure to publish is great, and they must be able to translate their discoveries into some kind of visual equation. Their images appear within a diverse range of media, including research journals, textbooks, popular science magazines, newspapers, television programmes and museum exhibitions, and each of these media have a different effect on the ways a representation works for its audiences. Science images that appear in the popular science media have the burden of attracting readers who may not yet have an interest in science.

Competence in science and mathematics appears to be diminishing, and better science images may help to counter the general public’s lack of interest in learning about science. Magazine editors and television producers have the added problem of selling their editorial agenda to advertisers. Images that are resolutely technical, or appear to lack popular appeal, may turn readers and benefactors away. Beyond the traditional media, science representations play an important role in contemporary culture by influencing legislation, jury decisions, medical opinions and capital investment. Science is a big business, where representations play a necessary role in encouraging stakeholder support for projects that require teams of researchers, state-of-the-art laboratories, new technologies and abundant cash.

The photographer Felice Frankel, a research scientist in the School of Science at the Massachusetts Institute of Technology (MIT), believes that designers have much to teach scientists about how to think visually. Frankel comes up with ways to amplify the visual values of what her colleagues – and often her colleagues only – see in the course of their research. As much as she relies on them for the arresting images she creates of their hidden worlds, she also winds up making them nervous.

In 2001, MIT scientists Todd Reynolds and Gerald Fink commissioned Frankel to photograph a three-inch wide colony of bakers’ yeast (saccharomyces cerevisiae) to document their research for the journal Science. Frankel photographed the yeast to emphasise its complex and unusual structure, known as a biofilm. Reynolds and Fink began studying the yeast to learn the ways that other micro-organisms, like fungi and bacteria, can stick to the plastic surfaces of prosthetic devices and fester into troublesome infections. They discovered that the innocuous yeast behaves similarly to pathogenic micro-organisms by becoming a biofilm itself. They hope that the yeast will help them to discover compounds that will arrest such infections.

Frankel’s contribution to the journal was significant. She was able to show detail in her depiction of the biofilm that the researchers were unable to capture themselves. After photographing the image, Frankel digitally removed the signs of the Petri dish that had contained the yeast, believing that it distracted her attention from the yeast’s signature floral patterns. Her ethereal image was selected as the journal’s cover image, a coup that would guarantee Reynolds and Fink wide attention for their research. Yet the scientists were not satisfied; they thought that Frankel’s editing eliminated their sample’s sense of scale, compromising its integrity.

Frankel thinks there are no easy solutions to resolve such controversies. ‘The scientific journals perpetuate their own visual traditions,’ she explains. ‘Scientists believe that if a research photograph or illustration is too refined, or too attractive, then it will not be taken seriously, and will sabotage their efforts to publish their work. It is never an easy sell, but scientific data can be informative and elegant.’

Frankel has made it her mission to get scientists and image-makers working more collaboratively. In June, she orchestrated a gathering of more than 175 designers and scientists at the J. Paul Getty Center in Los Angeles to debate contemporary problems in visualising science. The mit-sponsored forum, entitled ‘Image and Meaning 2: Discovering New Visual Expressions for Science and Technology’, provided a rare opportunity for the designers and scientists to share their experiences with diverse creative processes, gain greater fluency in using one another’s technologies, and develop innovative methods for explaining complex phenomena.

The event was the second in what Frankel expects to be a series; the first was held in 2001. [1] Participants at the forum represented every field of science imaginable, among them, biology, astronomy, medicine, robotics, satellite technology and computer science. The designers and image-makers in the audience, including animators, cartographers, film-makers, art directors and graphic designers, quickly realised that each individual field of science has its own peculiar culture, including language and identity patterns, which affects the way researchers make and use images within that field. Astronomers and molecular biologists, for example, may wish to remain aware of one another’s discoveries, yet they each have pictorial traditions and standards that do not easily translate between disciplines.

The structure of the forum allowed ample time for audience questions and reflections. The designers and scientists were grouped into workshops and were asked to resolve a series of problematic science images. The benefits of this structure quickly became obvious: scientists, as a whole, lack visual literacy, and do not have the artistic training to explain their research using the tools of text and image. Designers, on the other hand, lack specialised training in science yet must understand what they are representing before they can successfully visualise it. Throughout the presentations and workshops, various themes emerged that circumscribed what appeared to be forum’s most salient question: Is there a necessary boundary line that separates the cultures of design and science, and if so, where does science culture end and design culture begin?

Throughout the proceedings, the science community seemed to have difficulty understanding the need for text and image to function together. Text does not exist as a component of the physical world, and scientists appear to be uncomfortable incorporating typographic information into their research representations. Because each field of science has its own specialised vocabulary, the scientists appeared to favour image-based representations that can stand independently of labels or descriptions because they translate more fluidly across science’s cultural spectrum. In one workshop attended, Alyssa Goodman, a professor of astronomy at Harvard University, encouraged scientists to re-examine Galileo Galilei’s revolutionary manuscript Sidereus Nuncius (The Sidereal Messenger, 1610), in which he interspersed thumbnail-size sketches of Jupiter and its moons into his lines of explanatory text. ‘Imagine if we tried to get away with that in the journal Nature today,’ asked Goodman. ‘What kind of response would we get?’

However, the same image that allows an astronomer like Goodman to break new ground in astrophysics will likely fail as a tool for educating laypersons. Once a contemporary image succeeds as a tool for analysis, it must then be reworked as a tool for presentation, which requires a comprehensive understanding of the varying levels of interpretative authority that readers bring to their media. Science is commonly interpreted using metaphors, and the forum participants had much to say about the difficulties that arise from re-contextualising their data for popular consumption, or, to paraphrase the scholar William J. Mitchell, ‘getting the image to match the world.’ [2] As Kathy Stern, the graphic arts director of The New England Journal of Medicine observed, ‘Cast shadows, reflected light, perspective, surface texture and highlights all add useful layers of information for a medical practitioner working at a human scale, but do they make sense at a molecular level?’

Graphic designer Stefan Sagmeister presented examples of his experimental typographic work, encouraging scientists to follow his example and move away from the blandness of objectivity. Sagmeister favours images with subjectivity, and asserted that audiences can and do separate substance from style in whatever they read. The cartoonist (and former mathematician) Larry Gonick works at a more comfortable end of the accessibility spectrum by delivering science lessons in comic-book style, using humour and anthropomorphism to engage new audiences. Gonick’s work exemplified a narrative dimension that many of the forum attendees found to be simultaneously engaging and controversial, fearing that his molecular ‘characters’ lacked objectivity, and could mislead the public. Such work follows in the tradition of early geographers, who represented the trade winds as complacent beings puffing air from the boundaries of their projections. In line with such thinking, it is likely difficult for most people to look up into the night sky and locate a celestial object without seeing the compendium of beasts and heroes that informed our original understanding of the universe. Pictorial star charts may seem naive, yet constellations employ a metaphorical level of abstraction that works well for their inexperienced audience.

Pat Hanrahan, a professor of computer science and electrical engineering at Stanford University, believes that abstraction poses the greatest challenge for researchers in science today, especially those in his field. Many scientists have difficulty settling on an appropriate degree of accuracy, culling what they need to show from what they know. As Ben Fry, a lecturer on Visual and Environmental Studies at Harvard, said in his presentation, ‘It is easier to collect than to clarify.’

Scientists must often spend hundreds of hours staring at a single image to make sense of it. Complex images may be more comprehensive (and more correct to fellow scientists), but they can be difficult for non-experts to understand. Idealised representations are simpler, but they may also be incomprehensible to an audience that has no experience reading scientific shorthand. Frankel questioned whether the attendees, as a whole, might attempt to develop better tools for interpreting complexity that would work across the range of science disciplines.

Such an approach might work for the physical sciences. But in the biological sciences, the relationship between complexity and accuracy becomes increasingly difficult to ignore. Human factors add a cause-and-effect narration to scientific representation. Megan McGarrity works for Pollution Probe, a Canadian organisation that identifies and works on solutions for environmental problems such as water pollution and climate change. McGarrity considers physical (or quantitative) dimensions to be inseparable from narrative (or qualitative) dimensions. ‘Can an image help explain both the quantity and concentration of a given pollution emission,’ McGarrity asked, ‘as well as the environmental and health effects of that pollution at the community level?’

The term ‘dimension’, traditionally used to describe variables of space or time, was alternately presented throughout the forum as a term for describing the gamut of information variables. Following the example of the camera obscura, some representations function better as projections, that is, when multiple dimensions of visual information are compressed into a single plane. Following in the tradition of three-dimensional scientific models, like those that appear in many museums, other scientists felt the need to maximise the number of dimensions present within their representations. As the finale of his presentation, Dominique Brodbeck, a chief scientist and founding partner at Macrofocus, presented his company’s interactive visualisation of international purchasing power, entitled InfoScope, that distinguished between 42 unique dimensions, including population, net and gross wages, tax contributions, working hours per year, consumer prices and average worker incomes. Brodbeck’s interactive exemplified a common belief held by the scientists and designers alike: print has become a stifling medium for representing complexity. Many hoped that the Internet might soon replace or at least buttress the more traditional publishing media by giving both scientists and lay public greater control of what they wished to learn.

Throughout the forum, the subject that remained the least conclusive was the question of representing uncertainty. The project of investigation is never complete, and science representations must reconcile that which is known from that which is uncertain within the same image. Frankel faults science teaching: ‘Contrary to popular opinion, there is no black and white in science, just as in life,’ she said. ‘If we can begin to understand this as children, we would be better equipped to understand this as adults.’

How should scientists represent qualities such as randomness and abnormality, or that which is theoretical but inconclusive? Many scientists view ambiguity as evidence of insufficient research, yet there is much to learn about the physical world, and uncertainty can be a tool for encouraging further research and greater public curiosity.

The representations that we make today are likely to reward future scientists and image makers in ways we cannot yet anticipate. Alyssa Goodman, for her research in locating nascent stars in distant clouds of interstellar gas, is modifying a brain-imaging tool, the 3D Slicer, as a model for interdisciplinary thinking. [3] Goodman explains that such star-forming ‘tumours’ are difficult to detect without three-dimensional visualisation technology. ‘These new techniques are thrilling my astronomical colleagues,’ she said, ‘while exciting the medical imaging crowd about the widespread applicability of their work.’ Goodman’s presentation exemplified the forum’s spirit of collaboration, leaving little doubt that the cultural boundaries that separate scientists and designers are beginning to break down.

1. Anyone interested in participating in the next ‘Image and Meaning’ forum, scheduled for 2008, should contact [email protected].

2. William J. Mitchell, The Reconfigured Eye: Visual Truth in the Post-Photographic Era. Cambridge: mit Press, 1994; p.118.

3. The 3D Slicer was developed at the Brigham and Women’s Hospital’s Surgical Planning Laboratory by Goodman with fellow researchers Naomi Ridge and Michelle Borkin and imaging specialist Mike Halle.

Stuart McKee, designer, writer, San Francisco

First published in Eye no. 57 vol. 15.


Eye is the world’s most beautiful and collectable graphic design journal, published quarterly for professional designers, students and anyone interested in critical, informed writing about graphic design and visual culture. It is available from all good design bookshops and online at the Eye shop, where you can buy subscriptions, back issues and single copies of the latest issue.