Importantly, increase cognitive load on the frontal cortex, and afterward subjects become less prosocial*—less charitable or helpful, more likely to lie.46 Or increase cognitive load with a task requiring difficult emotional regulation, and subjects cheat more on their diets afterward.*47
So the frontal cortex is awash in Calvinist self-discipline, a superego with its nose to the grindstone.48 But as an important qualifier, soon after we’re potty-trained, doing the harder thing with our bladder muscles becomes automatic. Likewise with other initially demanding frontal tasks. For example, you’re learning a piece of music on the piano, there’s a difficult trill, and each time as you approach it, you think, “Here it comes. Remember, tuck my elbow in, lead with my thumb.” A classic working-memory task. And then one day you realize that you’re five measures past the trill, it went fine, and you didn’t have to think about it. And that’s when doing the trill is transferred from the frontal cortex to more reflexive brain regions (e.g., the cerebellum). This transition to automaticity also happens when you get good at a sport, when metaphorically your body knows what to do without your thinking about it.
The chapter on morality considers automaticity in a more important realm. Is resisting lying a demanding task for your frontal cortex, or is it effortless habit? As we’ll see, honesty often comes more easily thanks to automaticity. This helps explain the answer typically given after someone has been profoundly brave. “What were you thinking when you dove into the river to save that drowning child?” “I wasn’t thinking—before I knew it, I had jumped in.” Often the neurobiology of automaticity mediates doing the hardest moral acts, while the neurobiology of the frontal cortex mediates working hard on a term paper about the subject.
The Frontal Cortex and Social Behavior
Things get interesting when the frontal cortex has to add social factors to a cognitive mix. For example, one part of the monkey PFC contains neurons that activate when the monkey makes a mistake on a cognitive task or observes another monkey doing so; some activate only when it’s a particular animal who made the mistake. In a neuroimaging study humans had to choose something, balancing feedback obtained from their own prior choices with advice from another person. Different PFC circuits tracked “reward-driven” and “advice-driven” cogitating.49
Findings like these segue into the central role of the frontal cortex in social behavior.50 This is appreciated when comparing various primates. Across primate species, the bigger the size of the average social group, the larger the relative size of the frontal cortex. This is particularly so with “fission-fusion” species, where there are times when subgroups split up and function independently for a while before regrouping. Such a social structure is demanding, requiring the scaling of appropriate behavior to subgroup size and composition. Logically, primates from fission-fusion species (chimps, bonobos, orangutans, spider monkeys) have better frontocortical inhibitory control over behavior than do non-fission-fusion primates (gorillas, capuchins, macaques).
Among humans, the larger someone’s social network (measured by number of different people texted), the larger a particular PFC subregion (stay tuned).51 That’s cool, but we can’t tell if the big brain region causes the sociality or the reverse (assuming there’s causality). Another study resolves this; if rhesus monkeys are randomly placed into social groups, over the subsequent fifteen months, the bigger the group, the larger the PFC becomes—social complexity expands the frontal cortex.
We utilize the frontal cortex to do the harder thing in social contexts—we praise the hosts for the inedible dinner; refrain from hitting the infuriating coworker; don’t make sexual advances to someone, despite our fantasies; don’t belch loudly during the eulogy. A great way to appreciate the frontal cortex is to consider what happens when it is damaged.
The first “frontal” patient, the famous Phineas Gage, was identified in 1848 in Vermont. Gage, the foreman on a railroad construction crew, was injured when an accident with blasting powder blew a thirteen-pound iron tamping rod through the left side of his face and out the top front of his skull. It landed eighty feet away, along with much of his left frontal cortex.52
The two known pictures of Gage, along with the tamping rod.
Remarkably, he survived and recovered his health. But the respected, even-keeled Gage was transformed. In the words of the doctor who followed him over the years:
The equilibrium or balance, so to speak, between his intellectual faculties and animal propensities, seems to have been destroyed. He is fitful, irreverent, indulging at times in the grossest profanity (which was not previously his custom), manifesting but little deference for his fellows, impatient of restraint or advice when it conflicts with his desires, at times pertinaciously obstinate, yet capricious and vacillating, devising many plans of future operations, which are no sooner arranged than they are abandoned in turn for others appearing more feasible.
Gage was described by friends as “no longer Gage,” was incapable of resuming his job and was reduced to appearing (with his rod) as an exhibit displayed by P. T. Barnum. Poignant as hell.
—
Amazingly, Gage got better. Within a few years of his injury, he could resume work (mostly as a stagecoach driver) and was described as being broadly appropriate in his behavior. His remaining right frontal cortical tissue had taken on some of the functions lost in the injury. Such malleability of the brain is the focus of chapter 5.
—
Another example of what happens when the frontal cortex is damaged is observed in frontotemporal dementia (FTD), which starts by damaging the frontal cortex; intriguingly, the first neurons killed are those mysterious von Economo neurons that are unique to primates, elephants, and cetaceans.53 What are people with FTD like? They exhibit behavioral disinhibition and socially inappropriate behaviors. There’s also an apathy and lack of initiating behavior that reflects the fact that the “decider” is being destroyed.*
Something similar is seen in Huntington’s disease, a horrific disorder due to a thoroughly weird mutation. Subcortical circuits that coordinate signaling to muscles are destroyed, and the sufferer is progressively incapacitated by involuntary writhing movements. Except that it turns out that there is frontal damage as well, often before the subcortical damage. In about half the patients there’s also behavioral disinhibition—stealing, aggressiveness, hypersexuality, bursts of compulsive, inexplicable gambling.* Social and behavioral disinhibition also occur in individuals with stroke damage in the frontal cortex—for example, sexually assaultive behavior in an octogenarian.
There’s another circumstance where the frontal cortex is hypofunctional, producing similar behavioral manifestations—hypersexuality, outbursts of emotion, flamboyantly illogical acts.54 What disease is this? It isn’t. You’re dreaming. During REM sleep, when dreaming occurs, the frontal cortex goes off-line, and dream scriptwriters run wild. Moreover, if the frontal cortex is stimulated while people are dreaming, the dreams become less dreamlike, with more self-awareness. And there’s another nonpathological circumstance where the PFC silences, producing emotional tsunamis: during orgasm.
One last realm of frontal damage. Adrian Raine of the University of Pennsylvania and Kent Kiehl of the University of New Mexico report that criminal psychopaths have decreased activity in the frontal cortex and less coupling of the PFC to other brain regions (compared with nonpsychopathic criminals and noncriminal controls). Moreover, a shockingly large percentage of people incarcerated for violent crimes have a history of concussive trauma to the frontal cortex.55 More to come in chapter 16.
The Obligatory Declaration of the Falseness of the Dichotomy Between Cognition and Emotion
The PFC consists of various parts, subparts, and sub-subparts, enough to keep neuroanatomists off the dole. Two regions are crucial. First there is the dorsal part of the PFC, especially the dorsolateral PFC (dlPFC)—don’t worry about “dorsal” or “dorsolateral”; it’s just jar
gon.* The dlPFC is the decider of deciders, the most rational, cognitive, utilitarian, unsentimental part of the PFC. It’s the most recently evolved part of the PFC and the last part to fully mature. It mostly hears from and talks to other cortical regions.
In contrast to the dlPFC, there’s the ventral part of the PFC, particularly the ventromedial PFC (vmPFC). This is the frontocortical region that the visionary neuroanatomist Nauta made an honorary member of the limbic system because of its interconnections with it. Logically, the vmPFC is all about the impact of emotion on decision making. And many of our best and worst behaviors involve interactions of the vmPFC with the limbic system and dlPFC.*
The functions of the cognitive dlPFC are the essence of doing the harder thing.56 It’s the most active frontocortical region when someone forgoes an immediate reward for a bigger one later. Consider a classic moral quandary—is it okay to kill one innocent person to save five? When people ponder the question, greater dlPFC activation predicts a greater likelihood of answering yes (but as we’ll see in chapter 13, it also depends on how you ask the question).
Monkeys with dlPFC lesions can’t switch strategies in a task when the rewards given for each strategy shift—they perseverate with the strategy offering the most immediate reward.57 Similarly, humans with dlPFC damage are impaired in planning or gratification postponement, perseverate on strategies that offer immediate reward, and show poor executive control over their behavior.* Remarkably, the technique of transcranial magnetic stimulation can temporarily silence part of someone’s cortex, as was done in a fascinating study by Ernst Fehr of the University of Zurich.58 When the dlPFC was silenced, subjects playing an economic game impulsively accepted lousy offers that they’d normally reject in the hopes of getting better offers in the future. Crucially, this was about sociality—silencing the dlPFC had no effect if subjects thought the other player was a computer. Moreover, controls and subjects with silenced dlPFCs rated lousy offers as being equally unfair; thus, as concluded by the authors, “subjects [with the silenced dlPFC] behave as if they can no longer implement their fairness goals.”
What are the functions of the emotional vmPFC?59 What you’d expect, given its inputs from limbic structures. It activates if the person you’re rooting for wins a game, or if you listen to pleasant versus dissonant music (particularly if the music provokes a shiver-down-the-spine moment).
What are the effects of vmPFC damage?60 Lots of things remain normal—intelligence, working memory, making estimates. Individuals can “do the harder thing” with purely cognitive frontal tasks (e.g., puzzles where you have to give up a step of progress in order to gain two more).
The differences appear when it comes to making social/emotional decisions—vmPFC patients just can’t decide.* They understand the options and can sagely advise someone else in similar circumstances. But the closer to home and the more emotional the scenario, the more they have problems.
Damasio has produced an influential theory about emotion-laden decision making, rooted in the philosophies of Hume and William James; this will soon be discussed.61 Briefly, the frontal cortex runs “as if” experiments of gut feelings—“How would I feel if this outcome occurred?”—and makes choices with the answer in mind. Damaging the vmPFC, thus removing limbic input to the PFC, eliminates gut feelings, making decisions harder.
Moreover, eventual decisions are highly utilitarian. vmPFC patients are atypically willing to sacrifice one person, including a family member, to save five strangers.62 They’re more interested in outcomes than in their underlying emotional motives, punishing someone who accidentally kills but not one who tried to kill but failed, because, after all, no one died in the second case.
It’s Mr. Spock, running on only the dlPFC. Now for a crucial point. People who dichotomize between thought and emotion often prefer the former, viewing emotion as suspect. It gums up decision making by getting sentimental, sings too loudly, dresses flamboyantly, has unsettling amounts of armpit hair. In this view, get rid of the vmPFC, and we’d be more rational and function better.
But that’s not the case, as emphasized eloquently by Damasio. People with vmPFC damage not only have trouble making decisions but also make bad ones.63 They show poor judgment in choosing friends and partners and don’t shift behavior based on negative feedback. For example, consider a gambling task where reward rates for various strategies change without subjects knowing it, and subjects can shift their play strategy. Control subjects shift optimally, even if they can’t verbalize how reward rates have changed. Those with vmPFC damage don’t, even when they can verbalize. Without a vmPFC, you may know the meaning of negative feedback, but you don’t know the feeling of it in your gut and thus don’t shift behavior.
As we saw, without the dlPFC, the metaphorical superego is gone, resulting in individuals who are now hyperaggressive, hypersexual ids. But without a vmPFC, behavior is inappropriate in a detached way. This is the person who, encountering someone after a long time, says, “Hello, I see you’ve put on some weight.” And when castigated later by their mortified spouse, they will say with calm puzzlement, “But it’s true.” The vmPFC is not the vestigial appendix of the frontal cortex, where emotion is something akin to appendicitis, inflaming a sensible brain. Instead it’s essential.64 It wouldn’t be if we had evolved into Vulcans. But as long as the world is filled with humans, evolution would never have made us that way.
Activation of the dlPFC and vmPFC can be inversely correlated. In an inspired study where a keyboard was provided to jazz pianists inside a brain scanner, the vmPFC became more active and the dlPFC less so when subjects improvised. In another study, subjects judged hypothetical harmful acts. Pondering perpetrators’ responsibility activated the dlPFC; deciding the amount of punishment activated the vmPFC.* When subjects did a gambling task where reward probabilities for various strategies shifted and they could always change strategies, decision making reflected two factors: (a) the outcome of their most recent action (the better that had turned out, the more vmPFC activation), and (b) reward rates from all the previous rounds, something requiring a long retrospective view (the better the long-term rewards, the more dlPFC activation). Relative activation between the two regions predicted the decision subjects made.65
A simplistic view is that the vmPFC and dlPFC perpetually battle for domination by emotion versus cognition. But while emotion and cognition can be somewhat separable, they’re rarely in opposition. Instead they are intertwined in a collaborative relationship needed for normal function, and as tasks with both emotive and cognitive components become more difficult (making an increasingly complex economic decision in a setting that is increasingly unfair), activity in the two structures becomes more synchronized.
The Frontal Cortex and Its Relationship with the Limbic System
We now have a sense of what different subdivisions of the PFC do and how cognition and emotion interact neurobiologically. This leads us to consider how the frontal cortex and limbic system interact.
In landmark studies Joshua Greene of Harvard and Princeton’s Cohen showed how the “emotional” and “cognitive” parts of the brain can somewhat dissociate.66 They used philosophy’s famous “runaway trolley” problem, where a trolley is bearing down on five people and you must decide if it’s okay to kill one person to save the five. Framing of the problem is key. In one version you pull a lever, diverting the trolley onto a side track. This saves the five, but the trolley kills someone who happened to be on this other track; 70 to 90 percent of people say they would do this. In the second scenario you push the person in front of the trolley with your own hands. This stops the trolley, but the person is killed; 70 to 90 percent say no way. The same numerical trade-off, but utterly different decisions.
Greene and Cohen gave subjects the two versions while neuroimaging them. Contemplating intentionally killing someone with your own hands activates the decider dlPFC, along with emotion-related regions that respond to aversive
stimuli (including a cortical region activated by emotionally laden words), the amygdala, and the vmPFC. The more amygdaloid activation and the more negative emotions the participant reported in deciding, the less likely they were to push.
And when people contemplate detachedly pulling a lever that inadvertently kills someone? The dlPFC alone activates. As purely cerebral a decision as choosing which wrench to use to fix a widget. A great study.*
Other studies have examined interactions between “cognitive” and “emotional” parts of the brain. A few examples:
Chapter 3 discusses some unsettling research—stick your average person in a brain scanner, and show him a picture of someone of another race for only a tenth of a second. This is too fast for him to be aware of what he saw. But thanks to that anatomical shortcut, the amygdala knows . . . and activates. In contrast, show the picture for a longer time. Again the amygdala activates, but then the cognitive dlPFC does as well, inhibiting the amygdala—the effort to control what is for most people an unpalatable initial response.
Chapter 6 discusses experiments where a subject plays a game with two other people and is manipulated into feeling that she is being left out. This activates her amygdala, periaqueductal gray (that ancient brain region that helps process physical pain), anterior cingulate, and insula, an anatomical picture of anger, anxiety, pain, disgust, sadness. Soon afterward her PFC activates as rationalizations kick in—“This is just a stupid game; I have friends; my dog loves me.” And the amygdala et al. quiet down. And what if you do the same to someone whose frontal cortex is not fully functional? The amygdala is increasingly activated; the person feels increasingly distressed. What neurological disease is involved? None. This is a typical teenager.
Behave: The Biology of Humans at Our Best and Worst Page 6