Alex Soojung-Kim Pang, Ph.D.

I study people, technology, and the worlds they make

Month: September 2010 (page 1 of 4)

Resisting prediction in medicine

A few weeks ago I came across this article in Slate about how physicians don't do a good job of estimating how long terminally ill people have to live:

Doctors prefer not to prognosticate for three reasons: We don't like to be wrong; we don't want to take away hope for survival or good quality of life in the time that remains; and we just aren't adequately trained to do it. And our reluctance to make such guesses means that when we do try to predict the future, we're pretty lousy at it….

Since doctors typically avoid making predictions, these tools are infrequently dusted off and put to use. Our collective reluctance to offer patients a prognosis makes us less accurate in the rare instances we actually do it.

In his seminal book Death Foretold: Prophecy and Prognosis in Medical Care, Nicholas Christakis, a medical doctor and sociologist, argues that medical science has given the processes of diagnosis and treatment disproportionate emphasis in the educational curricula of doctors…. [A]voiding prognosis is a professional norm for doctors at all levels of training. In our research, teaching, and communication, we focus almost exclusively on the ever-expanding sciences of diagnosis and treatment, leaving prognosis almost entirely to the side.

Making predictions about the lives of… cancer patients can be particularly tricky. William Dahut, clinical director of the Center for Cancer Research at the National Cancer Institute, blames "a general lack of understanding of the specific biology of the cancer as well as a general lack of understanding of the biology of the individual." Doctors and scientists often refer to an individual's biology as "host factors," making allowances for the fact that patients are indeed different—in immunity, resilience, and attitude. The difficulty in accounting for such differences is another reason that predictive accuracy is so low….

Christakis argues that studying and delivering prognoses to patients is part of the ethical obligation of doctors to their patients. "Furthermore," he writes, "physicians should legitimate discussions regarding prognosis not only with their patients but with each other." As such, doctors would recast the professional norm to include open and frank discussion of prognosis in medical care.

In so doing, we need to strive for honesty and avoid "hanging crepe," the idea of delivering a poor prognosis simply to combat our tendency to be overly optimistic and to keep our hands clean: If the patient dies, I predicted it and therefore appear accurate; if the patient outlives my prediction, everyone is pleasantly surprised and thus I'm not held accountable.

We know thanks to Philip Tetlock how expert political judgment works. It would be interesting to look at a variety of different professions or disciplines that are under pressure to make predictions or forecasts, and see if there are interesting differences in the ways physicians, meteorologists, financial researchers, intelligence analysts, and others handle those demands.

links for 2010-09-28

  • Thesis exploring the intersection of design and futures.
    (tags: design future methodology paperspaces)
  • "We operate constantly in nano-, micro- and macro-future spaces. Nano-future predictions are made by our brain even though we are not consciously aware of them. Micro-futures, along with macro-futures comprise our personal observable and recognizable future horizon…. So how can we enhance our ability to recognize alternative futures, to see beyond day-to-day routines and familiar models? How could we use the information flows around us to explore our possible micro- and macro-futures? How to expand personal future horizon? We need to be actively bombarded with alternatives that challenge our models of subjective reality. Alternative futures can be exposed when we confront surprising things or recognize previously unnoticed factors of our everyday life."
    (tags: future prediction cognition personal_futures)
  • "FutureSelf is an evocative game-like system that constructs an interactive textual narrative describing the future of an individual. The evolving narrative can be treated as a future forecast but also as an alternate path of one’s existence. FutureSelf creates future simulations using the digitized information about the individual and her surroundings (e.g. social data). The system utilizes automatically the available information that is stored and processed in digital networked environments."
    (tags: simulation design future research personal_futures)
  • The aim of this study was to produce a validated satiety index of common foods…. The results show that isoenergetic servings of different foods differ greatly in their satiating capacities. This is relevant to the treatment and prevention of overweight and obesity.
    (tags: diet satiety)
  • "Studies by Australian researcher Dr. Susanne Holt at the University of Sydney have developed one of the most exciting diet concepts ever. Called the "Satiety Index", it was developed by having students come in the morning and eat 240-calorie portions of a specific food. Then they rated their feelings of hunger every 15 minutes, and over the next two hours, students could go to a buffet table and eat as much as the liked, all under the observation of researchers. Using white bread as the baseline of 100, they scored 38 different foods that were given to the students. Foods scoring higher than 100 were judged to be more satisfying than white bread, while those under 100 were less satisfying. Foods that have a higher satiety index keep hunger down longer, and would be better choices for those who want to lose weight." Winners: baked potato, fish, beans, beef, apples, oranges, oatmeal.
    (tags: diet)

Procrastination

Yep, this is it:

Procrastination from ism studios on Vimeo.

Moving R&D overseas

Ed Luce in the Financial Times worries about the shift of R&D in American (or American-founded) companies to China, and hopes that the departure of Larry Summers will occasion a rethinking of neoliberal trade policies.

Take Applied Materials, a big US manufacturing company, which earlier this year shifted its chief technology officer and research and development operations to China. The company said it needed its R&D to be close to the source of its manufacturing operations and to its biggest future market. This is the opposite of what is supposed to happen. America was meant to keep the high-end jobs at home, while China would get all the low-value added production.

When I was at IFTF I wrote about the gravitational pull of manufacturing, and argued that it's actually very hard to disaggregate R&D and product development (the high-end creative stuff that we were supposed to keep here in California) from manufacturing (which could be done by teenage girls and brown people in- well, who gives a damn where it's done). This is true for several reasons. Historically, some of the biggest innovations have emerged in the course of working on production problems (Bill Leslie's work explicates this relationship nicely). As designers know, creating something that can be mass-produced is not a trivial intellectual exercise: there's a special kind of genius necessary to make prototype (like the computer mouse, say) and turn it into a product. Manufacturing high-tech objects can be pretty damn hard, and requires more tacit knowledge and skill than we usually realize (Intel's Copy Exactly program is a great example of a company's attempt to get its hands around this fact). Finally, if a company has to choose between having imaginative research and getting products out the door, it's going to subordinate the former to the latter.

Not everyone sees this as a bad thing. Matthew Yglesias, for example, argues that

the alleged need for R&D to be proximate to manufacturing options (plausible) cuts in both directions. Conventional wisdom is that manufacturing operations will all drift to low-wage countries. But if the USA is a better location for R&D than China, and if it’s strongly desirable to co-locate R&D and manufacturing operations, then many firms will want to retain manufacturing operations in the United States of America. So if this story is right, then more and better education for America is the key to retaining high-wage manufacturing jobs.

I think this is wrong for two reasons. First, it's a lot harder to move the infrastructure for manufacturing than it is to move R&D, and it's very hard to get a factory back (or just as important, the skill necessary to manufacture things) once you've lost it. It's much easier to have production engineers fly to Taiwan or Shanghai every 6 weeks (though it sucks for those people). Second, it raises the question of why manufacturing jobs have left the U.S. at all, if R&D can attract manufacturing?

Are Post-its evil?

My friend Anthony Townsend turned me onto this entertaining rant against Post-Its and design thinking: Jamer Hunt argues that the Post-It filled wallboard has become a wrongheaded symbol of creativity.

The predominant image of design in the 21st century is that cliché of the empty conference room or studio-just after some feverish brainstorming extravaganza-plastered with Post-it notes … as if the act of design had suddenly morphed into some strange game of pin the Post-it on the mind map. How is it possible that the wonderfully complex process of design has devolved to the point that we now commonly represent it by the leftover artifacts of quickie ideation? Is that all there is?

As someone who's written pretty extensively on the use of paper media and paper spaces in collaborative creative work, naturally I was intrigued by Hunt's argument, but he seems a lot more concerned about the Post-it as symbol of design. He doesn't seem to be arguing that it's a tool that leads you to do bad work, or is too weak to support good work. So what's wrong with the Post-it as symbol?

The problem is that in serving as a substitute for the whole of design, the Post-it represents only a small fraction of what makes design uniquely effective. It papers over the fact that ideation without materialization is not design. Designers discover as they turn ideas into thing (even when those things have no physical form). We gain true insight in the act of making a mark on a page or pushing pixels on the screen. We don't need to over-hype those processes, but to ignore them means that we shortchange the practice of design. Clever ideas are a dime-a-dozen-about the cost of Post-its.

Fair enough. The idea that ideas are what matters, and that the actual thing is kind of an afterthought is one of the scourges of our age. This kind of creative Platonism downplays (or just misunderstands) the difficulty of actually making good things, and the role that solving production problems can play in innovation. (I was turned onto this by two very different sources: Bill Leslie's work on Bell Labs and Western Electric, and Matthew Crawford's Shop Class as Soul Craft) To the degree that those of us in the delirious professions treat brainstorming (and its material/visual expression) as an end-point rather than a stage in a bigger process, we sell ourselves short, and do a disservice to our clients.

links for 2010-09-24

  • "Cyber security experts say they have identified the world's first known cyber super weapon designed specifically to destroy a real-world target – [by attacking the automated control systems at] a factory, a refinery, or just maybe a nuclear power plant. The cyber worm, called Stuxnet, has been the object of intense study since its detection in June. As more has become known about it, alarm about its capabilities and purpose have grown. Some top cyber security experts now say Stuxnet's arrival heralds something blindingly new: a cyber weapon created to cross from the digital realm to the physical world – to destroy something…. Stuxnet is essentially a precision, military-grade cyber missile deployed early last year to seek out and destroy one real-world target of high importance – a target still unknown."
    (tags: security software internet endofcyberspace)
  • "Former executives and regulators, testifying Thursday before the Financial Crisis Inquiry Commission, all tried to shift blame for the giant company's problems…. If there was an underlying consensus, it was this: the financial crisis was either entirely unforeseeable or should have been spotted first by somebody else."
    (tags: unintended_consequences business)
  • As a nation, we cannot afford to live with Mr. Greenspan’s way of thinking. The truth is, he should have seen what was coming and offered a sober, apolitical warning. Everyone would have listened; when he talked about the economy, the world hung on every single word…. And at this point there is no reason to reflexively dismiss the analysis of those who foresaw the crisis. Mr. Greenspan should use his substantial intellect and unsurpassed knowledge of government to ascertain and explain exactly how he and other officials missed the boat. If the mistakes were properly outlined, that might both inform Congress’s efforts to improve financial regulation and help keep future Fed chairmen from making the same errors again.
    (tags: business unintended_consequences)
  • "The must read commentary of the weekend is a Sunday NYT OpEd by Michael Burry — he’s featured in Michael Lewis The Big Short — titled "I Saw the Crisis Coming. Why Didn’t the Fed?" Its another assault on Greenspan’s tarnished legacy: “ALAN GREENSPAN, the former chairman of the Federal Reserve, proclaimed last month that no one could have predicted the housing bubble. “Everybody missed it,” he said, “academia, the Federal Reserve, all regulators.” But that is not how I remember it. Back in 2005 and 2006, I argued as forcefully as I could, in letters to clients of my investment firm, Scion Capital, that the mortgage market would melt down in the second half of 2007, causing substantial damage to the economy."
    (tags: business unintended_consequences)
  • Nobody could have predicted that the bank’s highest-rated collateralized debt obligations — created by repackaging mortgage bonds into new securities — would lose so much money, Prince said. The chief risk officer didn’t understand the risks, nor did Citigroup’s senior traders and bankers, he said.
    (tags: business unintended_consequences)

links for 2010-09-23

links for 2010-09-22

  • "[G]irls’ education may be the single most cost-effective kind of aid work. It’s cheap, it opens minds, it gives girls new career opportunities and ways to generate cash, it leads them to have fewer children and invest more in those children, and it tends to bring women from the shadows into the formal economy and society. It’s not a panacea, of course. Lebanon and Sri Lanka were leaders in girls’ education, and both ended up torn apart by conflict. In India, the state of Kerala has done a fine job in girls’ education, but its state economy is still a mess and dependent on remittances. But overall, educating girls probably has a greater transformative effect on a country than anything else one can do."
    (tags: education development)
  • "[E]ducating girls may be the single most cost-effective way to empower and modernize societies, to help people help themselves. Larry Summers said back when he was chief economist at the World Bank that girls’ education was such a great investment that the question was not whether we could afford to educate girls, but whether we could afford not to. Of course, it’s also important to educate boys, but there are a couple of reasons why girls’ education brings an even higher return. First, it reduces birth rates very considerably, which brings the country a demographic dividend. Second, more educated people tend to have higher incomes, and women are more likely than men to use extra income to educate their own children and to start small businesses."
    (tags: education development)

links for 2010-09-21

  • (tags: unintended_consequences iraq warfare strategy)
  • In the light of the outstanding importance of hydrocarbons for global energy, the controversy over peak oil has become both pressing and emotionally charged…. The shaping of the future energy policy is presently based on modeling results and geological considerations only. We show that the existing predictions of the energy crisis are increasingly mixed-up with value-judgments. The value analysis of those forecasts allows us to suggest that at least part of the estimations are implicit reflections of predictors’ ends and values, and do not demonstrate a real ability to anticipate future conditions. Paradoxically, the question of oil reserves depletion is better understood when predictions are viewed as an instrument to impose the predictors’ values and intervene in the currently bustling oil market. The intervention in the oil prices may occur in either direction becoming a tool to justify values rather than an instrument for the acquisition of knowledge."
    (tags: forecasting energy oil peakoil)

links for 2010-09-20

  • The paper discusses the necessity for futures studies and argues the need for methods giving explicable understandings of future possibilities so that decisions and policies can be as future-proof as possible. A taxonomy for futures methodologies based on their passive, preventive or anticipatory characteristics is proposed. The anticipatory methodologies are further categorised into subjective and numerical approaches. The paper goes on to review some of the principal numerical approaches such as system dynamics and econometric methods. The subjective approaches, such as the extended scenario, Delphi and Field Anomaly Relaxation are considered and it is concluded that, in general, they, and especially Field Anomaly Relaxation, are the more fruitful line of attack on the futures problem. Some directions for further research are suggested.
    (tags: future methodology)
  • "I think that people should be held accountable for their predictions. One hopes they will be embarrassed enough to refrain from wasting our time and trees on future predictions."
    (tags: future forecasting accuracy)
  • Look at the debacles of the early 2000s- Iraq, Fannie Mae, Citigroup, Bernie Madoff- "it's amazing how much in common they all have. And not just in how they began but in how they ended: with those responsible being amazed at what happened, because…who could have known? Well, to paraphrase James Inhofe, I'm amazed at the amazement. In fact, when historians look for a name that sums up the Bush II years, they could do worse than calling them The "Who Could Have Known?" Era. Each of the disasters listed above was entirely predictable. And, indeed, was predicted. But those who rang the alarm bells were aggressively ignored, which is why it's important that we not let those responsible get away with the "Who Could Have Known?" excuse."
    (tags: unintended_consequences politics)
  • "In an interview Thursday with the AP, Vice President Cheney neatly summarized the failed Bush presidency. Comparing the financial meltdown and implosion of the American economy with the 9/11 attacks, Cheney insisted, "I don't think anybody saw it coming." As it turns out, from 9/11, sectarian conflict in Iraq and the election of Hamas to the Bush recession and the drowning of New Orleans by Hurricane Katrina, the leading lights of the Bush administration claimed they never saw it coming. Call it the "Nobody Could've Predicted Presidency.""
    (tags: unintended_consequences uncertainty prediction)
  • "Secondhand smoke poses risks to children, particularly those from low socioeconomic backgrounds. Recently, there has been an increase in tobacco-control policies designed to reduce children's exposure to secondhand smoke, including interventions to change parental smoking behaviors. However, little attention has been paid to understanding potential unintended consequences of such initiatives on mothers who smoke. As such, the objectives of this paper are to explore the potential consequences of tobacco-control policies designed to reduce children's exposure to secondhand smoke on socially disadvantaged mothers who smoke and to provide recommendations for research, policy, and practice…. Stigmatization research suggests that such policies may have unanticipated outcomes for socially disadvantaged mothers who smoke, such as decreased mental health; increased use of cigarettes or alcohol; avoidance or delay in seeking medical care; and poorer treatment by healthcare professionals."
    (tags: unintended_consequences boomerang_effect psychology public_health policy)
  • Some warnings and other public health interventions have been found to produce effects opposite to those intended. Researchers employing a variety of methods have observed these boomerang effects in connection with interventions in a number of different contexts. One possible explanation for such boomerang effects lies in the theory of psychological reactance, roughly defined as the state of being aroused in opposition to perceived threats to personal choice. In particular, some consumer reactions described in research on alcoholic beverage warnings, alcohol education efforts, and the minimum drinking age can be concisely explained in terms of psychological reactance. An obvious implication is that boomerang effects should be taken into account as one of the potential costs of launching a mass communication campaign or requiring a warning.
    (tags: unintended_consequences boomerang_effect policy)
  • Debate over the influence of postwildfire management on future fire severity is occurring in the absence of empirical studies. We used satellite data, government agency records, and aerial photography to examine a forest landscape in southwest Oregon that burned in 1987 and then was subject, in part, to salvage-logging and conifer planting before it reburned during the 2002 Biscuit Fire. Areas that burned severely in 1987 tended to reburn at high severity in 2002, after controlling for the influence of several topographical and biophysical covariates. Areas unaffected by the initial fire tended to burn at the lowest severities in 2002. Areas that were salvage-logged and planted after the initial fire burned more severely than comparable unmanaged areas, suggesting that fuel conditions in conifer plantations can increase fire severity despite removal of large woody fuels.
    (tags: unintended_consequences forestry environment policy)
  • In an alarming trend over the last few years, large swaths of farms, forests and wildlands permanently protected for the benefit of the public have been targeted for the siting of energy infrastructure projects. As climate and energy bills move through Congress, the push for rapid development of low carbon energy and new transmission lines should not, as an unintended consequence, undo years of work and public and private investment in conservation.
    (tags: unintended_consequences forestry environment policy)
  • Biodiversity is important in ecosystems and for the provision of ecosystem services including climate regulation. It can therefore play an important role in reducing climate change and its impacts, and protecting and improving societal wellbeing. However, there is growing concern that efforts to address climate change may have the unintended consequence of exacerbating biodiversity loss, and so reduce future options for responding to climate change.
    (tags: unintended_consequences environment policy climate)
  • Growing corn, converting it to ethanol, and transporting it utilize so much fossil fuel that the process generates nearly as much CO2 as simply burning gasoline in automobiles. The process also requires large amounts of water, which is becoming an increasingly scarce global resource as aquifers are over-pumped to meet existing irrigation needs. Demand for biodiesel in Europe has led to the unintended consequence of contributing to deforestation in Indonesia and Malaysia as people clear trees to grow palm oil plantations, which, among all the alternatives, produces the highest yield of biodiesel per acre.
    (tags: environment climate policy unintended_consequences)
  • Blames food shortages on government policies supporting biofuels.
    (tags: unintended_consequences climate environment)
  • More forests, deserts and grasslands in the U.S. will be used to produce energy under a proposal to cap greenhouse gases, an unintended consequence of efforts to fight global warming, according to a Nature Conservancy report…. Less land will be needed to grow corn for cleaner-burning ethanol and to support electric-generating wind turbines if legislation gives carbon-dioxide emitters more options to reach targets, said the report, published today in the online journal PloS One. Greater energy conservation can also reduce the amount of land needed for development…. “In the scenarios we considered, there is a tendency for greater reductions in greenhouse-gas emissions to be associated with a greater total new area affected by energy development,” the report said. “A decrease in U.S. emissions increases the new area impacted, although the magnitude of the effect is policy-specific.”
    (tags: policy environment climate unintended_consequences)
  • Sarbanes-Oxley "set out to restore investors’ confidence in financial markets by improving corporate governance. However, in the case of at least one group of companies, the act seems to have produced unexpected results. A study co-authored by Thomas Lys (Professor of Accounting Information & Management at the Kellogg School of Management) indicates that the managements of poorly governed foreign-domiciled firms responded to the act by closing shop in the United States. They continued operations—replete with their managerial faults—in their own countries. “So with the act we have exported the problem,” Lys says. “This is an unintended consequence.”"
    (tags: finance accounting unintended_consequences policy)
  • A brilliant, tragic example of unintended consequences: "The decision to go or to stay also depended not least on people's experiences in their immediate surroundings. "Seen from the point of view of Auschwitz," the editors write in their introduction, "a tragic insight opens up: the more openly anti-Semitic the 'Ayran' neighbours, customers, and co-workers were at the beginning of Nazi rule, the faster the victims were able to take the decision to flee and ultimately save their lives. If their Christian acquaintances and friends were friendly and helpful, the persecuted were more likely to opt to stay, thus cutting their chances of survival dramatically." Seen from the point of view of Auschwitz, this bitter pill tells us, the yardstick of moral judgement is shattered and enemies of Jews inadvertently save Jewish lives and friends of Jews become their gravediggers."
    (tags: unintended_consequences history)
  • Long-range, political-economic forecasting cannot be appraised in terms of empirically demonstrated accuracy. Yet as ‘scientific research programs’, futures studies can be assessed in terms of methodological dependability and progressive problem shifts. The methodology of developmental constructs meets these criteria; policy debates and international conflicts can be viewed as competitions among developmental sequences, which progress best, if cast as provisional rather than as general laws. Developmental constructs can incorporate historical lessons without the rigidity of single, dominant analogies. The approach applies with equal robustness to long-term economic growth futures studies and international conflict and mediation.
    (tags: forecasting prediction future accuracy)
Older posts

© 2017 Alex Soojung-Kim Pang, Ph.D.

Theme by Anders NorenUp ↑