Science of Stress, Testosterone & Free Will | Dr. Robert Sapolsky

The episode opens by introducing Dr. Robert Sapolsky as a professor of biology and neurosurgery at Stanford University whose laboratory has studied stress and hormones, including testosterone and estrogen. His work is described as combining primatology, including field studies, with human behavior, in an effort to understand how humans, as Old World primates, are shaped by both biology and psychology.

The introduction lists his popular books, "Why Zebras Don't Get Ulcers," "The Trouble with Testosterone" and "Behave: The Biology of Humans at Our Best and Worst," and notes that he is close to finishing a new book, "Determined: The Science of Life Without Free Will." The stated agenda covers stress and its conscious and subconscious control over us, testosterone, estrogen and hormone replacement therapy, and how these influence psychology and social interaction. The introduction also warns that the interview was recorded remotely, so listeners may hear occasional audio artifacts.

The first question asks how to think about short-term and long-term stress, since the stress response keeps us alive and can sharpen mental acuity. The answer describes two graphs. On the first, short-term stress brings all sorts of benefits: it saves you from the predator, or helps you think more clearly and focus during a presentation. Once stress becomes chronic, the curve goes "downhill from there." The disputed question is how long it takes for short-term stress to become long-term.

That threshold is called "somewhat arbitrary." Still, the chronic stressors most people face, such as years of daily traffic jams or an abusive boss, are described as "undeniably in the chronic range."

Our goal is not to cure people of stress, because if it's the right kind, we love it.

The second graph treats the right amount of stress as stimulation. People pay to be stressed by scary movies or rollercoasters. Function peaks at an optimal level of stimulation. Below it, function drops into what we call boredom. Above it, function drops into what we call stress. That optimum is described as the target everyone aims for.

The discussion notes that the physiological stress response looks very similar to excitement about a positive event. One speculation is that a neuromodulator such as dopamine might explain the difference. The follow-up question asks what biology tells us about valence, the quality that makes an experience feel awful or exhilarating, and where valence resides.

The answer stays at a mechanical level. When a situation makes your heart race, speeds your breathing and uses your muscles, the brain-activation profile is roughly the same whether the event is wonderful or terrible. The one exception is the amygdala. If it is part of the activation, the experience counts as adverse, and the more the amygdala is involved, the more it biases you toward interpreting the situation as even more awful.

The amygdala in some ways is the checkpoint as to whether we're talking about excitement or terror.

The amygdala leads into testosterone. One participant recalls hearing that the amygdala is among the brain regions most densely packed with testosterone receptors. The response places the amygdala at the center of both the fear-versus-aggression and excitement-versus-stress axes. It then argues that "almost everybody out there has a completely wrong idea" about what testosterone does.

The common view is built on real correlations. In virtually every species, males have more testosterone and are more aggressive. Seasonal breeders show testosterone surges during territorial fighting season. Castrating any mammal, including humans, lowers aggression. From this it seems easy to conclude that testosterone causes aggression. The discussion rejects that conclusion and says the evidence shows up both in behavior and in the amygdala.

It lowers the threshold for the things that would normally provoke you into being aggressive so that it happens more easily.

The behavioral evidence comes from a primate example. Put five male monkeys together and they form a dominance hierarchy, with number one at the top and number five miserable. Give number three large doses of testosterone and he gets into more fights. On closer inspection, though, he does not challenge numbers one and two. He keeps deferring to them ("brown-nosing them") exactly as before and becomes "a miserable terror" to numbers four and five. In this account, testosterone amplifies existing patterns of aggression and the social learning behind them.

The same logic applies at the neural level. Testosterone does not make amygdala neurons fire spontaneously or generate fear and aggression on their own. If the amygdala is already being stimulated, testosterone increases the firing rate and, as mentioned in passing, shortens the after-hyperpolarization. The conclusion is that testosterone's effects cannot be described outside the social context that determines how testosterone-related behaviors are treated.

The theme there exactly is it's not creating aggression, it's just upping the volume of whatever aggression is already there.

Asked whether relative testosterone levels track status in a hierarchy, the answer is yes, "but in a way that winds up being totally uninteresting." Older endocrinology texts carried two reliable findings: higher testosterone predicts more aggression and more sexual activity. A closer look reverses much of the causal arrow. Sexual behavior and aggression raise testosterone levels, while levels measured beforehand are described as "barely predictive" of what happens next. (A brief interruption by a jingling dog is welcomed on the show.)

For explaining differences between individuals, the argument goes, these levels "don't matter a whole lot." A researcher could spend a career on social circumstances that produce three and a half percent more circulating testosterone and expect interesting consequences that do not materialize. Testosterone is described as "somewhat of a yes or no modulator" of much subtler social factors that are already in place.

The hypothetical of raising someone's testosterone by 30%, male or female, gets the reply "Essentially zero effect at all," because the brain is not that sensitive to fluctuations in testosterone levels. On the reverse direction, the discussion notes that even watching your favorite team play, sitting in an armchair with potato chips, can raise testosterone if you are devoted enough. The point is that the rise is driven by "the psychological framing" of aggression, not its physicality.