August in San Diego 8. Biology, Light, and Aesthetics
Neuroscience For Architecture, Urbanism & Design
Michael A. Arbib
This is the eighth of a series of nine posts on the A«Nblog reporting on the “Neuroscience For Architecture, Urbanism & Design” Intersession held at NewSchool of Architecture & Design in San Diego on August 12-15, 2019. The individual posts range in length from 1300 to 3000 words. The first post provides an overview of the series, along with a Table of Contents with links to each of the posts. A PDF of the whole series may be found here.
In addition to the visual mechanisms of the brain that support visual perception of the world around us(see below), the brain has mechanisms sensitive to light that help adjust our circadian rhythms. Satchin Panda studies the diverse circadian rhythms that relate to sleep, nutrition, hormones, and how these related rhythms are disrupted by electrical light, with a host of possible adverse effects on the downtime that diverse bodily systems need, and that sleep provides. He found that many effects of light related to circadian rhythms persist in mice even in the absence of the rods and cones that provide the retinal receptors implicated in visual perception. He discovered melanopsin – a novel photo pigment in just a few thousand cells in the mammalian retina, hardwired to the master clock in the brain, more sensitive to blue light than other parts of the spectrum and thus less sensitive to the candlelight and firelight that our ancestors relied on at night. This clearly has implications both for lifestyle and for the way architects employ artificial light, especially during night time. Studies show that intermittent versus continuous lighting may be similar in effect on circadian rhythms, but one needs to start dimming lights 2 to 3 hours before bedtime.
Fred Marks discussed how light plays a crucial part in our lives and how architecture makes use of this. This includes celebratory purposes and seasonal timing, as at Stonehenge. Light is projected into the Pantheon in Rome via the oculus. Light is a symbol of knowledge and understanding and, yes, enlightenment. The Phoenix Public Library (Will Bruder) celebrates the summer solstice every June. Medieval castles were built with thick masonry walls that, for defensive purposes, contained only small apertures. Changes in European building practice from the Norman to the Gothic, such as flying buttresses, allowed walls to be thinner and allowed churches to include large stained glass windows. Glass windows in the renaissance brought more light through the walls, cast iron provided new ways of realizing this in the 19th century and, even more dramatically, steel girders in the 20th century removed weight support from the walls, which could then contain more glass and admit more light. One may then add a scrim to reduce the light, but how much is needed? Artificial light progressed from candles to gas, then electricity and now LEDs. Now more people stay awake late into the night, and 20% are on night shifts.
Now light can be distributed to create engaging displays. Marks cited the Olafur Eliason installation, the Weather Project, at the Tate Modern in London in 2003. Representations of the sun dominated the expanse of the Turbine Hall and engaged many people, exemplifying the socialization induced by light.
Many traditional farmers are out in the fields all day with bright light exposure except when going into the house for a nap or a meal. Panda now offers an App to help people monitor their light exposure, and finds that the built environment offers very different amounts of light. He measured his exposure to bright light in different venues and found that his office at the Salk offered surprisingly little exposure. Bright light that works well in the day is counter-productive experienced in, e.g., a 24 hour supermarket at night. With Marks, he analyzes the different light exposure of school children in different settings in San Diego. What about controlling warm versus blue lighting in the classroom? Well, a typical room has 4 sources of light, natural, overhead, task, and screens, so it is challenging to assess their relative contributions to the overall effect.
The mammalian brain has two main pathways that link vision with perception (and much more). One travels from the retina to the superior colliculus in the midbrain and plays a key role in rapid involuntary eye movements, the other is the thalamocortical system that travel from retina via thalamus to cerebral cortex. Tom Albright offered a quick tour of the thalamocortical system. Information reaching primary visual cortex then passes to diverse interacting visual areas of cerebral cortex including those involved in depth and motion processing and color, and then onward to object recognition – an assemblage of specialized processes collectively yielding visual experience of the world. David Hubel and Thorsten Wiesel showed that there are cells in primary visual cortex that act as “edge detectors,” firing most actively when an edge or gradient of a particular orientation appears at a particular location in the visual field.
Albright offered a two-part theory concerning the early stages of visual processing:
- The visual environment has measurable statistics. For example, angular correlation between contours decreases with the distance between them. Color correlation also declines, but in a different way. The statistics are different for a natural scene and for, e.g., a Jackson Pollock picture.
- Organizational properties. Nearby features are represented nearby in the brain. Hubel and Wiesel’s edge detectors are arranged in “pinwheels” – the preferred orientation of cells changes progressively around the pinwheel, while nearby pinwheels link to nearby locales in the visual field. Anatomical connections link similar features. [Caveat: While nearby features are represented nearby in the brain, this breaks down in later regions.]
Later regions of the visual system resolve the visual input by interpreting parts of the input as objects and recognizing the relations between them or the actions that are taking place. The bottom-up processing of the retinal input is complemented by top-down influences (often unconscious ones) based on hypotheses and expectations shaped by prior experience. In short, memory has a great impact on bottom-up processing — a continuum from pure stimulus to pure imagery (and even hallucination). To illustrate this, Albright showed an image that most of the audience could not interpret. He then “changed their brains forever, without their permission” by showing the image of a bearded man from which the first image had been abstracted. Thereafter, one could not look at the first image without seeing the man’s face – and I can attest that this was the case for me even a year after first experiencing this effect.
Tom Albright notes that the visual system evolved to detect relevant patterns in the environment, and continued his discussion by asking “What makes beauty.” His first answer is that the Fay Jones chapel in the woods and the Notre-Dame rose window and cable-stay bridges are all seen as beautiful because of regular changes in angle that relate to the properties of the low-level visual system. [However, a not-so-low-level system is crucial in that this beauty is based on relations between contours, not local Hubel-Wiesel features. How does one get from local edges to perceived contours in the visual scene? Computational models suggest that nearby neurons related to features consistent with a continuous contour tend to excite each other, while those that do not tend to inhibit each other (via inhibitory neurons), thus maintaining the activity that defines a possible contour consistent with the retinal stimulation.]
To build on the above examples, Albright cited E.H. Gombrich on “the sense of order.” Gombrich contrasts
- Easy adjustment(familiarity) in perceiving a visual scene. Examples include regular tessellations and mandalas. Albright sees the Sydney Opera House as having conventional beauty based on easy adjustment.
- Easy arousal(the sense of novelty). For this, he stresses the impact of memory on the bottom-up processing. A continuum from pure stimulus to pure imagery. The pattern requires assessing different hypotheses to impose sense on what you see. It requires “perceptual improvisation.”
Easy arousal for Albright is the type of beauty one sees in non-representational art. He relates this to the notion of William James of “the victorious assimilation of the new” but does acknowledge Robert Irwin’s contrary view that such perception is “just Rohrschaching it.”
I note, however, that our perception of any visual scene could be assessed in terms of easy adjustment versus easy arousal. Such a classification holds whether the scene is beautiful, ugly or just blah. Thus the dichotomy may be part of a theory of beauty, but does not answer the motivating question “What makes beauty?” It is also interesting to note that Albright’s account is based on “disembodied vision,” whereas Mallgrave approached Einfühlung in terms of embodied cognition, empathy and mirror neurons. Developing a new account that benefits from both these approaches poses a fine challenge for NfA.
Albright is conducting with Sergei Gepshtein and a Baltimore architect a study of kids responding to patterns in a classroom, finding those which can reduce stress, or improve academic performance. He gave a talk on this at AIA, suggesting how NfA may improve the built environment for educating children, improving health, etc.
Alison Whitelaw commented that Albright told us about vision, novelty ,and familiarity and then asked, “How do we trade these off in design?” E.O. Wilson speculates on the comfortable level of complexity. What is the impact of environments that support easy arousal and easy exploration? Consider increasing architectural complexity in the approach to an art gallery, but reducing it in the display galleries to provide a foil to support attention to the complexity of the artworks. Whitelaw has applied similar ideas to school design, to a public utility building (with public inclusion), and to a high school performing art center.
Marcos Novak took us on a wild ride in his talk “Enacting civilization. Neuroaesthetics and thermodynamics of beauty,” starting on a musical note by playing Roxy Music’s Every dream home a heartache: “Standard of living is rising daily. … penthouse perfection, but what goes on, what to do there? Better to pray there … bungalow ranch style, all of its comforts seem so essential … Inflatable doll, my breath is inside you …” He asked how we can make things to “enact beauty.” Why do some civilizations build beautiful cities and others do not? Novak briefly explored diverse scientific theories. Let me note a number of his observations and ideas.
He builds on notions of thermodynamics and Boltzmann’s explanation via statistical mechanics.
A diagram of clock hand, all rounded at the center, narrowing to a point (the narrow PhD thesis) symbolized that everybody’s good at something but no one is minding the shop (e.g., global warming and catastrophe). But the sea urchin has many spikes, and each spike is not an end in itself, but one of many that contribute to the animal’s living.
For an exhibition of his work in an Eisenman museum, he was faced with a 65 foot blank wall, but had no time to fill it with things. Overnight, he generated a program to define shapes in a limited supply of vinyl and specify how to cut them from the sheet and how to position each piece on the wall. What about an algorithm for placement in a a3D space, an algorithm where people can react in a shared virtual environment?
World making, world shaping, relating building to poesis. His transLAB seeks to make things from which new species emerge. What then is a species if not defined by Darwin’s reproductive fitness? Mediated worlds. Beyond STEM and STEAM to incorporate the humanities and making. An array that’s more than n-dimensional in which one can locate efforts. Novak listed a breathless array of topics, from quantum mechanics to poetry to 20th century citizenship!
On to beauty, which is above wisdom. Beauty is rare and it is difficult to achieve. But there are places where a community could create beauty together, as in Kyoto. Why did Athens spend half its military budget to build the Parthenon, a building that could last 2500 years? Maybe they were building for the future. So how can we build a beautiful civilization and save the world before we destroy it? The core meaning of aesthetics. is just “sensation.” It is the door, not the grand hall!
John Zeisel. Warren Neidich. John Onians and his neuroarthistory. Neurocognitive poetics. The Evil Robots of the Ancient World – see the book Gods and Robotsby Adrienne Mayor.
We are deeply invested in expanding the scope of the human; expanding the degrees of freedom. Artificial gene synthesis, adding letters to the bases AGTC. Thermodynamics. Fourier analysis. Signal analysis. Building the city. Maximizing rent income is not the right constraint. If everybody has freedom, will the result be harmonic or beautiful.
Why does time move in one direction? Why is entropy increasing? We are all unique. But unique does not imply equiprobable. Beauty is rare. Beauty is not increasing freedom, it must balance the freedom with constraints. A ladder from the superficial to the just, and the just is difficult. Life and consciousness are involved with the transcendent but also the material world. Novak’s final image was of Brancusi’s studio.
Leach questioned the claim that beauty is rare. We can see a beautiful sunset almost every day. Novak responded that few planets have beautiful sunsets. Tourists flock to beautiful places and destroy them in the process. But they could go home and ask architects to design beautiful buildings for them. Leach responded that gay men can talk about beauty but perhaps others are too manly to talk about it!!
Q: On her first day visit to Iceland, she was struck by the beauty of the sunset. On her second visit, in summer and with flocks of tourists, she did not find the sunset beautiful. Can neuroscience explain this? Novak responded that perhaps the first had rarity value that the second did not. [But this cannot be quite right. We can see sunset after sunset as beautiful. Perhaps the point is that if sunsets contain enough variety of shape and color, they invoke the easy arousal of which Albright spoke.] Gepshtein added that it’s early days for neuroaesthetics. To move forward, we need to understand the dynamic processes of appreciation in relation the impact of the new. [Recall the TV series of Bob Hughes on modern art, “The Shock of the New,” and the mention of William James by Albright.] Novak closed the discussion by stressing that novelty does not guarantee beauty – car manufacturers come up with new exterior designs to sell more cars, not in the quest for beauty.
About Michael A. Arbib
Michael Arbib is a pioneer in the study of computational models of brain mechanisms, especially those linking vision and action, and their application to artificial intelligence and robotics. Currently his two main projects are “how the brain got language” through biological and cultural evolution as inferred from data from comparative (neuro)primatology, and the conversation between neuroscience and architecture. He serves as Coordinator of ANFA’s Advisory Council and is currently Adjunct Professor of Psychology at the University of California at San Diego and a Contributing Faculty Member in Architecture at NewSchool of Architecture and Design. The author or editor of more than 40 books, Arbib is currently at work on When Brains Meet Buildings, integrating exposition of relevant neuroscience with discussions of the experience of architecture, the design of architecture, and neuromorphic architecture.