Q: When might your car be willing to kill you?

A: Consider this situation: Driving down a narrow mountain road and rounding a sharp curve, you suddenly encounter a group of pedestrians in your path. Should you sacrifice yourself by steering over the cliff or save yourself by plowing through them?

As self-driving cars take to the roads, such moral decisions will inevitably arise, argue psychological scientist Jean-Francois Bonnefon et al. in "Science" magazine. Using a series of carefully crafted Internet surveys, the authors investigated U.S. public opinion regarding autonomous vehicles (AVs). When asked if AVs should "minimize casualties on the road" versus "protect the driver at all costs," respondents overwhelmingly supported so-called "utilitarian" AVs programmed to sacrifice occupants for the greater good. Yet when asked if they would actually purchase a utilitarian AV or support government regulations requiring that AVs be utilitarian, the response was typically "no."

Yet AVs are eventually expected to reduce traffic fatalities dramatically (by 90 percent according to some estimates), so the greater good may actually be served by building non-utilitarian AVs since they may more likely be purchased. As the authors observe, "Figuring out how to build ethical autonomous machines is one of the thorniest challenges in artificial intelligence today."

Q: In 2000, one-quarter of the global population was near-sighted, requiring some form of corrective lens. By 2050, that percentage is projected to be: a. slightly higher, b. 35 percent, c. 50 percent, d. 65 percent.


A: According to Australia's Brien Holden Vision Institute, almost half the world will be near-sighted (answer c), reports Diana Kwon in "Scientific American" magazine. Once believed to be largely genetic in origin, myopia "is, in fact, a socially determined disease," states ophthalmology researcher Ian Morgan. Though many believe that reading and staring at computer screens are to blame, there's little evidence to support this idea. More likely it's that "people, especially children, spend too little time outside — a handful of studies show that lack of sunlight exposure from long periods indoors correlates with myopia." In fact, countries with higher GDP tend to have higher rates of myopia, perhaps because those with higher socioeconomic status often spend less time outdoors.

This suggests a possible intervention: "A recent trial revealed that children who spent an extra 40 minutes outside each day for three years were less likely to become myopic than those who did not."

Q: In recent decades, DNA testing has exonerated hundreds of people convicted of serious crimes, such as rape and murder. According to the Innocence Project, what are the most common reasons for these wrongful convictions?

A: Inaccurate eyewitness testimony accounts for more than 70 percent of such convictions. Studies have repeatedly shown that human memory is much less reliable than we typically assume.

More than 25 percent of those wrongfully convicted make a false confession or incriminating statement, often during aggressive interrogation. Plus, 15 percent involve statements from people given incentives, such as plea deals to testify, sometimes without being disclosed to the jury.

Also, many forensic techniques, including hair and bite mark analysis and shoe print comparisons, are unreliable. Adding to the bad science is misapplication of statistics. Finally, inadequate defense at trial or on appeal and misconduct of law enforcement demonstrate a criminal justice system clearly administered by human beings with all their foibles.

Send STRANGE questions to brothers Bill and Rich at sbtcolumn@gmail.com.