A Monumental, but Flawed, Effort to Understand Behavior

Reynold Spector

Behave: The Biology of Humans at Our Best and Worst. By Robert M. Sapolsky. New York: Penguin Press, 2017. ISBN 978-1594205071. 790 pp. Hardcover, $35.00.

In his long (790 page), extensively referenced book Behave: The Biology of Humans at Our Best and Worst, neurobiologist and primatologist Robert Sapolsky attempts to uncover the cause and consequences of many aspects of human behavior employing techniques and results from neuroscience, evolution, psychology, sociology, molecular biology, genetics, and moral philosophy with a sociological bent. He begins his analysis and synthesis of behavioral data with the underlying neurobiological and hormonal causes of behavior. Later he focuses on fetal development, childhood, adolescence, and adulthood. Still later, he focuses on cultural and ecological factors that influence behavior. He spends considerable time on morality and less on religion.

In many places, Sapolsky plunges into controversial areas. For example, he argues that Steven Pinker’s thesis that “people have gotten less awful” over the centuries is overly optimistic. He implies that Pinker is a Pangloss. Sapolsky points out that when you take time (duration) as well as population size into account, he finds that World War II, the An Lushan Rebellion (China) and the killings in World War I, the Taiping Rebellion (China), and Tamerlane’s depredations lead the list of horrific events. He also emphasizes that the recent horrific wars in Vietnam, Afghanistan, and Iraq; the horrendous killing crimes of Stalin and Mao; and the ongoing wars in Syria, Iraq, Afghanistan, and Africa disconfirm Pinker’s optimistic thesis. In another part of the book, he accuses Pinker of “cherry-picking.” (Pinker responds to such criticisms in a chapter of his new book Enlightenment Now that was published in the May/June 2018 Skeptical Inquirer.)

In other parts of Sapolsky’s book there are clear deficiencies or wrong statements. For example, he implicitly assumes the critical importance of the Western liberal notion of the primacy of harm and fairness, which in turn leads to the evolutionary theory of kin selection and reciprocal fairness. However, he ignores important non-Western attitudes that motivate behavior in the majority of humans: these include: 1) societies in which ingroup/outgroup dynamics and loyalty are crucial (e.g., Japan); 2) societies where authority, respect, and obedience are all important (neo-Confucian societies); 3) societies where spiritual purity is paramount; and 4) societies where noncarnal, nonviolent behavior is critical (Amish). Thus, in his discussion of morality, Sapolsky’s implicit assumptions lead to an overvaluing of certain behaviors, e.g., empathy (see Paul Bloom’s 2016 book Against Empathy: The Case for Rational Compassion for a contrary view). I, like many others, would argue you cannot understand many behaviors and events unless you take nonliberal non-Western worldviews into account. Sapolsky might be surprised to know that in Korea (where I served in the U.S. Army in 1969 and spoke Korean), Good Samaritans were widely viewed by Koreans as fools for giving away their family or government’s wealth to strangers.

Now, to the heart of Sapolsky’s book: What causes behavior and from whence does it originate? And what are the consequences? He focuses on “good” and “bad” behaviors and, at great length, on the underpinnings of “moral” behavior. However, one critical aspect of his discussions is missing: the thorough and systematic work of centuries of moral philosophers beginning with Plato, Aristotle, and several Chinese scholars culminating in the work of Bernard Williams and Derek Parfitt of Cambridge University. He also does not emphasize the critical role that the law plays in determining behavior, especially in America.

Sapolsky rightly emphasizes the central nervous system—especially the brain and, later in the book, hormones. However, he is apparently not familiar with (or disregards) reports that do not support his views. For example, he ignores nutrition and many other important environmental factors. He tries to localize where in the brain certain behaviors originate. He assumes brain scanning studies are meaningful, i.e., changes in brain scans associated with certain behaviors or perceptions are scientifically sound. However, he does not emphasize that brain scans, e.g., fMRI and PET scans, actually measure blood flow, blood volume, deoxygenated hemoglobin, or glucose and/or oxygen uptake; he does not note that non-neural processes can affect fMRI and PET. (See Jerome Kagan’s 2017 book Five Constraints on Predicting Behavior for a detailed critique of Sapolsky’s methods.)

In the many psychological studies he references, Sapolsky is not aware of or ignores the findings that many such studies are nonreplicable and that many are done in artificial laboratory conditions. Moreover, even if correct, such studies are often done on college volunteers and may not be generalizable to the broader population. Context greatly matters, as Kagan emphasizes.

Sapolsky also places too much emphasis on research employing mice, rats, primates, and particularly his favorites, baboons. The behavior and physiology of these animals often does not extrapolate to humans. A rat brain weighs 1 gram; a human brain 1 kilogram. Moreover, humans have neural structures, connections, and cortical areas that other mammals, including primates, do not have.

Other methodological problems abound throughout the book. For example, Sapolsky discusses almost all studies in terms of means (averages) and statistical significance. He rarely discusses magnitudes of such differences—or that in fMRI very small differences can be greatly magnified and overemphasized. This lack of focus on the magnitude of differences instead of statistical significance is a grievous fault.

Finally, I note two errors of fact in areas in which I am thoroughly familiar and have published. First, on page 560, Sapolsky states categorically that substance P is involved in depression. He says “drugs that block the action of substance P can have marked anti-depressant properties.” This is not true. In the 1990s when I was head of development at Merck, we hypothesized that a “blocker” of substance P in the brain would have anti-emetic and antidepressant properties. The former is true, and our drug aprepitant, now generic, a substance P blocker, was approved by the FDA almost two decades ago for use as an anti-emetic. However, although one initial small pilot study suggested that aprepitant had antidepressant properties, large controlled trials did not confirm this hypothesis. Thus, Sapolsky is wrong about substance P and depression.

More important, on pages 147–150 Sapolsky argues that in humans “throughout adult life there is neurogenesis in brain,” for example, there is the birth of new neurons in the hippocampus and cerebral cortex. This is an extremely important issue for neurobiology, pharmacology, exercise physiology, aging, and hormonal action. For example, in rats, exercise and antidepressants were shown to stimulate neurogenesis. Moreover, Sapolsky argues that Professor Pasko Rakic (Yale) was incorrect in doubting many reports of the role of neurogenesis in animals and humans. Because of Rakic’s negativity about neurogenesis in adult humans, Sapolsky makes an ad hominem attack on Rakic, indirectly suggesting Rakic held up the “field” for ten years. In fact, many of the earlier studies that reported neurogenesis in the adult human brain made errors. First, the authors of these studies assumed DNA synthesis in the human brain was in part due to the birth of new neurons. This turned out to be not correct. I worked in the field of DNA synthesis in brains of animals and humans for a decade. It is true that DNA synthesis in adult human brain neurons occurs, but not for neurogenesis (the birth of new neurons). DNA synthesis in neurons, as in other cells, serves several purposes, including DNA repair, the repair of the removal of the sixth base in DNA (hydroxymethylcytosine), and intracellular mitochondrial replication. Mitochondria last only about thirty days and need to be replaced frequently. DNA synthesis also occurs in degenerating and dying neurons.

A highly publicized report twenty years ago by Elizabeth Gould of neurogenesis in the adult primate neocortex turned out to be a claim that could not be replicated. Moreover, there were other kinds of errors in the early reports of widespread brain neurogenesis in animals as well as adult humans, e.g., mistaking dividing endothelial and glial cells for neurons. (Sapolsky quotes the initial incorrect or misleading reports.) Consequently, in the past few years, there has developed a consensus that in only one small part of the adult human brain (the dentate gyrus of the hippocampus) might there be neurogenesis. However, more recently, multiple reports using increasingly sophisticated techniques have not found neurogenesis in the adult human brain, or if there is, it is below the level of detection—much less than 1 percent per year in the dentate gyrus and nowhere else. (See Shawn F. Sorrells et al.’s 2018 article “Human Hippocampal Neurogenesis Drops Sharply in Children to Undetectable Levels in Adults” in the journal Nature.) Thus, these recent more sophisticated experimental investigations give good reasons to doubt that quantitatively important human adult neurogenesis occurs. Moreover, unlike in rodents, the belief that human depression is due to deficient neurogenesis is obviously incorrect and that exercise, environmental enrichment, antidepressants, etc., might enhance neurogenesis in adult human brains is also almost certainly incorrect since significant neurogenesis in adult humans doesn’t occur. Sapolsky’s ad hominem attacks on Pasko Rakic deserve an apology.

To explain the lack of neurogenesis in adult humans, in 1985 Rakic and others proposed a theory that suggests that the lack of human neurogenesis is beneficial, so that long-term memory and other key functions can last over the entire life span. As I’ve noted, the extant reliable data support this very important theory. The amount of time, effort, and taxpayer money wasted on murine models is staggering.

I believe that Sapolsky is biased in his discussion of neurogenesis. He is trying to show there is tremendous “plasticity” in the brain that can be affected by environmental factors. This is undoubtedly true; for example, nutrition, education, and experience can affect the development and function of the human brain and attendant behavior. But there is no convincing evidence that after about age seven neurogenesis contributes much if anything to plasticity.

In summary, this book is a monumental effort to describe and understand behavior, especially human behavior. But in many places it is deficient—with an implicit point of view that affects Sapolsky’s analysis and synthesis, many errors of fact and interpretation, and the lack of coverage of several key drivers of behavior, for example the law.

Reynold Spector

Reynold Spector, MD has served as a professor of medicine (and pharmacology and/or biochemistry) at Iowa, Stanford, and Harvard-MIT. He is currently clinical professor of medicine at the Robert Wood Johnson Medical School (New Jersey) and is the author of almost 200 peer-reviewed scientific papers and one textbook. His award-winning work has concerned itself principally with vitamin function, transport, and homeostasis in the central nervous system, the effect of food on the function of the kidneys, and the treatment of the poisoned patient. Dr. Spector also served as executive vice president in charge of drug development at Merck from 1987 to 1999, where he oversaw the introduction of fifteen new drugs and vaccines


Behave: The Biology of Humans at Our Best and Worst. By Robert M. Sapolsky. New York: Penguin Press, 2017. ISBN 978-1594205071. 790 pp. Hardcover, $35.00. In his long (790 page), extensively referenced book Behave: The Biology of Humans at Our Best and Worst, neurobiologist and primatologist Robert Sapolsky attempts to uncover the cause and consequences …

This article is available to subscribers only.
Subscribe now or log in to read this article.