The Molecule of More

ence and Technology, 1(2), 51–63.
Friedman, T. (Producer), & Jones, P. (Director). (1996). NOVA: Einstein Revealed. 
Boston, MA: WGBH.
Kuepper, H. (2017). Short life history: Hans Albert Einstein. Retrieved from 
http://www.einstein-website.de/biographies/einsteinhansalbert_content.
html
James, I. (2003). Singular scientists. Journal of the Royal Society of Medicine, 96(1), 
36–39. 

---

---

145
Chapter 5
POLITICS
Why we can’t just get along.
In which we learn how superpowers and hand 
sanitizers affect our political ideology
THE AUTHORS REGRET . . .
In April 2002 the American Journal of Political Science published a research 
report, “Correlation not Causation: The Relationship Between Per-
sonality Traits and Political Ideologies.” It was written by a group of
researchers from Virginia Commonwealth University who studied the 
link between political beliefs and personality traits. They found that the 
two were connected, and that the connection could be attributed to 
genes. Along the way, they noticed that certain personality traits were 
associated with liberals and others with conservatives.
They were particularly interested in a collection of personality 
features—what psychiatrists call a personality constellation—called P. The 
Conservative: A statesman who is enamored of existing evils, as 
distinguished from the liberal, who wishes to replace them with others.
—Ambrose Bierce, 
The Devil’s Dictionary

---

146
THE MOLECULE OF MORE
authors noted that people with low P scores are more likely to be “altru-
istic, well socialized, empathic, and conventional.” By contrast, people 
who have high P scores are “manipulative, tough-minded, and practi-
cal,” and present characteristics such as “risk-taking, sensation-seeking, 
impulsivity, and authoritarianism.” They concluded, “As such, we expect 
higher P scores to be related to a more conservative political attitude.”
What they predicted was exactly what they found. The stereotypes, 
they said, were true: conservatives tend to be impulsive and authoritarian 
while liberals tend to be well socialized and generous. But in science, when 
you find just what you expect, it can be a red flag. And in January 2016, 
fourteen years after the original report, the journal published a retraction:
The authors regret that there is an error in the published 
version of “Correlation not Causation: The Relationship 
Between Personality Traits and Political Ideologies.” The 
interpretation of the coding . . . was exactly reversed.
Somebody had flipped the labels. The correct interpretation was the 
opposite of what they reported. It was the liberals in their study—not 
the conservatives—who were manipulative, tough-minded, and prac-
tical. And it was the conservatives, not the liberals, who tended to be 
altruistic, well socialized, empathic, and conventional. Many people 
expressed surprise at this reversal. But if we consider what the study 
found at its most basic level and how it relates to the dopamine sys-
tem, the revised results make good sense—certainly more sense than 
the original findings, which were widely heralded but exactly backward.
THE LIMITATIONS OF 
PERSONALITY MEASURES
Psychologists have worked for decades to develop ways 
to measure personality. They found that personality can be 
divided into different domains, such as how open a person is 

---

147
POLITICS
to new experiences or how self-disciplined he is. American 
psychologists divide personality into five domains, while the 
British prefer three. Either way, when a scientist focuses on 
one of the domains, she is measuring only a slice of a per-
son’s personality, not the whole person. Consider two nurses 
who both have high compassion scores. At first glance, one 
might imagine two similar people. But there are other per-
sonality domains as well. One nurse might be outgoing and 
emotional, while the other is introverted and restrained. 
Even though nurses may have some personality features in 
common, they are a group made up of unique individuals.
Another limitation of personality measurements is that 
scientists usually report a group’s 
average
score. So if a 
study finds that liberals are more risk-taking than conser-
vatives, it’s likely that within that group of liberals there are 
some who are safety-seeking. Studies of personality help us 
predict what a group of people will do, but they are less 
helpful in predicting what an individual will do.
PROGRESSIVES IMAGINE A BETTER FUTURE
The characteristics the study eventually associated with liberals— 
risk-taking, sensation-seeking, impulsivity, and authoritarianism—are 
the characteristics of elevated dopamine.
1
But do dopaminergic people 
1 In fact, a group of scientists from the Institute of Psychiatry in London found that 
dopamine receptors were crowded together more tightly in the brains of people 
with high 
P
scores compared to those with lower scores. Dense receptor pack-
ing led to stronger dopamine signals, which in turn led to the emergence of the 
distinctive personality features. The connection is also seen when we look at what