Scientific American - 09.2019

Graphic by Jen Christiansen September 2019, ScientificAmerican.com 77

SOURCE: “INDIVIDUALS WITH GREATER SCIENCE LITERACY AND EDUCATION HAVE MORE POLARIZED BELIEFS ON CONTROVERSIAL SCIENCE TOPICS,” BY CAITLIN DRUMMOND AND BARUCH FISCHHOFF, IN

PROCEEDINGS OF THE NATIONAL

ACADEMY OF SCIENCES USA,

VOL. 114, NO. 36; SEPTEMBER 5, 2017

thought the technology had a greater potential to

cause widespread catastrophes. That pattern held

true for other technologies as well.

To find out whether knowing more about a tech-

nology changed this pattern, we asked technical ex-

perts the same questions. The experts generally

agreed with laypeople about nuclear power’s death

toll for a typical year: low. But when they defined risk

themselves, on a broader time frame, they saw less po-

tential for problems. The general public, unlike the ex-

perts, emphasized what could happen in a very bad

year. The public and the experts were talking past

each other and focusing on different parts of reality.

UNDERSTANDING RISK

DID EXPERTS always have an accurate understanding of

the probabilities for disaster? Experts analyze risks by

breaking complex problems into more knowable

parts. With nuclear power, the parts might include

the performance of valves, control panels, evacuation

schemes and cybersecurity defenses. With GMO crops,

the parts might include effects on human health, soil

chemistry and insect species.

The quality and accuracy of a risk analysis depend

on the strength of the science used to assess each part.

Science is fairly strong for nuclear power and GMOs.

For new technologies such as self-driving vehicles, it is

a different story. The components of risk could be the

probability of the vehicle laser-light sensors “seeing” a

pedestrian, the likelihood of a pedestrian acting pre-

dictably, and the chances of a driver taking control at

the exact moment when a pedestrian is unseen or un-

predictable. The physics of pulsed laser-light sen sors

is well understood, but how they perform in snow and

gloom is not. Research on how pedestrians interact

with autonomous vehicles barely exists. And stu dies of

drivers predict that they cannot stay vigilant enough

to handle infrequent emergencies.

When scientific understanding is incomplete, risk

analysis shifts from reliance on established facts to ex-

pert judgment. Studies of those judgments find that

they are often quite good—but only when experts get

good feedback. For example, meteorologists routinely

compare their probability-of-precipitation forecasts

with the rain gauge at their station. Given that clear,

prompt feedback, when forecasters say that there is a

70  percent chance of rain, it rains about 70  percent of

the time. With new technologies such as the self-driv-

ing car or gene editing, however, feedback will be a long

time coming. Until it does, we will be unsure—and the

experts themselves will not know—how accurate their

risk estimates really are.

THE SCIENCE OF CLIMATE SCIENCE

EXPERT JUDGMENT , which is dependent on good feed-

back, comes into play when one is predicting the costs

and benefits of attempts to slow climate change or to

adapt to it. Climate analyses combine the judgments

of experts from many research areas, including obvi-

ous ones, such as atmospheric chemistry and ocean-

ography, and less obvious ones, such as botany, ar-

chaeology and glaciology. In complex climate anal-

yses, these expert judgments reflect great knowedge

driven by evidence-based feedback. But some as pects

still re main uncertain.

My first encounter with these analyses was in 1979,

as part of a project planning the next 20 years of cli-

When the Public Disagrees

about Science

'³Ç ̧§žîž`D§§ā` ̧³îß ̧þxßäžD§ä`žx³îž‰`žääøxäjÇ ̧§DߞąDîž ̧³žäßxDîxßD­ ̧³

Uxîîxߝž³… ̧ß­xlÇx ̧ǧxͳþxäDî ̧ßääDÿäx†x`îž³îÿ ̧³Dîž ̧³D§äøßþxāä

ž³îšx7Í3Í5šxäøßþxāäj` ̧³lø`îxlž³öććéD³löć¿ćj` ̧­Už³xlî ̧` ̧þxߥøäî

­ ̧ßxîšD³éjŠććÇx ̧ǧxÍ0Dßîž`žÇD³îäÿxßxDä¦xlÿšDîîšxāUx§žxþxl ̧³äxþxßD§

š ̧îî ̧Ǟ`äD³lÿšxîšxßîšxāDßxxlÿžîšä`žx³îž‰`` ̧³äx³äøäÍäxlø`Dîž ̧³

D³lä`žx³`x§žîxßD`āž³`ßxDäxlD­ ̧³§žUxßD§äD³l` ̧³äxßþDîžþxäjä ̧lžlîšxžß

lžþxߐx³`xÍ5šžä­DāUxUx`Døäx­ ̧ßxÿx§§þxßäxlÇx ̧ǧxDßxUxîîxßDîîø³xlî ̧

îšxÇ ̧äžîž ̧³ ̧…îšxžßÇ ̧§žîž`D§ß ̧øÇD³l­ ̧ßx` ̧³‰lx³îž³lx…x³lž³žîÍ

Government should support stem cell research

The universe developed from the big bang

General Education Science Education Science Literacy

Less More

More consistent with

å`Ÿy ́``¹ ́åy ́åùå

Less consistent with

å`Ÿy ́``¹ ́åy ́åùå

Less More Less More

Political Identity

Extremely

liberal

Liberal

Slightly

liberal

Moderate

Slightly

conservative

Conservative

Extremely

conservative

Humans developed through evolution

Climate change is a serious concern