Beauty & Pain

I thought I'd share this brief item from a mid-September issue of the New Scientist with you. I think there are implications for HCI, and a connection to value sensitive design. 

In pain? Take one masterpiece, three times a day
18 September 2008
From New Scientist Print Edition. Subscribe and get 4 free issues.

THE power of art to heal emotional wounds is well known, but could contemplating a beautiful painting have the same effect on physical pain?

To investigate, Marina de Tommaso and a team from the University of Bari in Italy asked 12 men and women to pick the 20 paintings they considered most ugly and most beautiful from a selection of 300 works by artists such as da Vinci and Botticelli.

They were then asked to contemplate either the beautiful paintings, or the ugly painting, or a blank panel while the team zapped a short laser pulse at their hand, creating a pricking sensation.

The subjects rated the pain as being a third less intense while they were viewing the beautiful paintings, compared with contemplating the ugly paintings or the blank panel. Electrodes measuring the brain's electrical activity suggested a reduced response to the pain when the subject looked at beautiful paintings (Consciousness and Cognition, DOI: 10.1016/j.concog.2008.07.002).

While distractions are known to reduce pain in hospital patients, de Tommaso says this is the first result to show that beauty plays a part. "Hospitals have been designed to be functional, but we think that their aesthetic aspects should be taken into account too," she says.

The Human Brain - With one hundred billion nerve cells, the complexity is mind-boggling. Learn more in our cutting edge special report.

From issue 2674 of New Scientist magazine, 18 September 2008, page 14

It would be interesting to design an experiment along these lines for an interface. 

a) present the test group with 20 interfaces, let them pick the most beautiful and the ugliest

b) assign each individual to focus on a blank screen, a beautiful interface, or an ugly one until a prompt appears

c) instead of a pain-inducing laser, use a 'critical failure' message, a 'something very bad has happened message'

d) ask the user to rate the level of pain/annoyance

Then determine whether the 'beautiful' interface moderates the effects of the error message ... or something along these lines. I bet you can think of a better experiment along these lines than I have! ~Matt Rolph

Is this a beautiful interface? Is the technology it serves as a gateway to more frustrating than average? Less? About the same? If an interface is beautiful, can that moderate frustration (presuming that what can go wrong always will, that there will be something frustrating about using even the best design)? Image from: http://www.physorg.com/newman/gfx/news/iphone_inhand.jpg

Value-Sensitive Design

"Value Sensitive Design and Information Systems" in Human-Computer Interaction in Management Information Systems: Foundations (Friedman, B., Kahn, P. H., Jr., & Borning, A., 2006) brings to mind an important concern with value-sensitive design. One of the scenarios examined was that of using a small camera mounted on the side of a building to provide a live feed of a public space exterior to the building. The article details an extensive study of so-called indirect stake-holders in this situation, the people who moved within view of the camera. These subjects expressed concern with their privacy and the ethical implications of this perceived violation thereof.

A careful examination of the relevant figures (16.2) shows that the camera was mounted above existing building windows. Further, the view pictured showed that the employees were provided with a view scaled to be equivalent to that from one of the windows near the camera. Because of the limitations of the screen and the human eye, they must necessarily be unable to see more detail than from the corresponding window. Functionally, then, this system must invade privacy less than or only as much as the windows in the building. Naturally, windows on buildings facing public spaces are not ethically concerning in this social context. The subjects, then, are not ethically impaired by the system, despite their belief that they can be. There are two possible reasons for these concerns to be warranted. First, if the video system records and saves the stream, this distinguishes the technology from windows. However, there is no indication that this is the case. Second, the inability to see back into the camera as one can do with a window might also make these situations different. However, one-way glass is common in offices, especially when windows are large. This one-way glass is viewed essentially the same as two-way glass, leading this to also be an invalid mechanism for differentiation.

This highlights the need for a good understanding of the technologies on the part of all parties involved before useful discourse can be achieved.

Sustainable interaction design: Invention & disposal, renewal & reuse (Blevis, E., 2007) examines the concept of sustainable interaction design. Though the concepts therein were largely sound, one particular oversight was very distracting---open-source development. The greatest concerns raised in the paper were rapid obsolescence of hardware supporting interactions. Open-source eases this problem in several respects.

First, open development usually means "competitive" products can borrow whatever back-end code is necessary to support the other's hardware (for example, the widely-used networking BSD stack). If the shortcomings of one product become too onerous, one of its contemporaries is likely to be capable of drop-in replacement without substituting any hardware.

Second, whether because of a smaller, elite developer base (compare the city population-sized team of Microsoft programmers to the relatively tiny handful of core developers for Linux plus GNU plus KDE), open code reviewing, or some other aspect of open source, software upgrades engender mandatory hardware updates much less often in open-source solutions than their proprietary counterparts. One famous example is KDE. Numerous major updates have lowered, not raised, the system requirements, including the 40% reduction in memory requirements from the 3.x series to the 4.x series (See system requirements on http://docs.kde.org/); this is even in spite of massive increases in functionality with the 4.x series! By contrast, Windows Vista was widely criticized for not adding enough new functionality (http://en.wikipedia.org/wiki/Criticism_of_Windows_Vista), but it pushed up minimum hardware requirements across the board, often substantially.

A game design methodology to incorporate social activist themes (Flanagan, M., & Nissenbaum, H., 2007) expands on a topic of Sustainable interaction design: Invention & disposal, renewal & reuse, social activism, in the context of gaming. Though the paper is interesting, it fails to address the most central pre-condition to its relevance: what evidence exists that game activities and morality translate to real life? Without this link, efforts to create ethically enhanced games are not only wasted, but counter-productive by drawing resources from veritably effective campaigns. The oft-studied topic of violence in video games has yet to draw a conclusive link between that game morality and real life, though evidence is mounting (for more, see http://en.wikipedia.org/wiki/Video_game_controversy). If this deeply cathartic or powerfully desensitizing (depending on ones position) form of interaction is still disputed, how can one treat the far more subtle topic of the efficacy of positive moral exposure in games as a forgone conclusion?

The section "Human values, ethics, and design" in The Human-Computer Interaction Handbook: Fundamentals, Evolving Technologies and Emerging Applications, 2nd Edition. (Friedman, B., & Kahn, P. H., Jr. , 2007) raises several interesting questions about the nature of morality in its pursuit of ethical perspective for HCI. The most fundamental, of course, is the question of moral epistemology. In attempting to address this (as yet, unsolved) question, the authors invoke an alleged cross-cultural moral similarity to construct what amounts to an argumentum ad populum. This, of course, is a fallacy. Worse, it is an unnecessary problem to approach. Morality isn't the relevant concern---ethics are. The apparent conflation of the two terms leads the authors to unnecessarily feel they must resolve issues of morality.

Illustrations of the danger of addressing personal values, morals, with social values, ethics, can be seen in the unpredictability of moral judgments. For example a recent study (citation unavailable, source was read several years ago) discovered variability in personal responses to a moral dilemma with equal ethical values. Subjects were asked to imagine several scenarios in which causing the death of one stranger saved a larger group of strangers from their death. Depending on the abstraction the subject had between their actions and the death of the individual, they were variously willing to perform the necessary action. Personally pushing the person over a cliff was often repulsive, but pulling a switch that caused the death was far more palatable. These reactions varied among subjects, but all subjects recognized the ethical equivalency of their actions.

A corresponding effect seems inevitable in software. Though personal morality may vary in some conditions, people seem more likely to be offended by decisions that deviate from their morals to align with someone else's morals than if the deviation from their morals were to align with socially-accepted ethics.