Our then ten-year-old son, David, was in the lineup at Venice Beach. Most of the waves were in the six-foot range, with faces about twice his height. Sitting next to him on my board, I could see a hundred people dispersed along the beach, jockeying for the location where the best waves would begin to break. With luck, we could surf ahead of the break for about fifteen seconds before jumping off our boards or twisting back out to sea to catch another. David positioned himself correctly and paddled into a large, surfable wave. After a long drop down the face, he shot across ahead of the break before turning into the wave, leaping over it, and paddling back out. It was turning into one of his best days of surfing.
Then a large, more powerful set of waves came plowing toward the beach. Surfers began paddling hard to clear them so they would not get smashed by a wave or have to struggle through the whitewater. I paddled out as quickly as I could over the first of several eight-foot waves. David and many others were not as lucky. To my horror, he did not make it over the first wave and disappeared into the foaming white impact zone. As I bobbed up on the next large wave, I saw David in the foamy detritus of the previous wave. The leash that connected his ankle to the board had broken. Along with his board, he lost his ability to paddle quickly out of danger.
I bobbed up on the next wave and, with increasing dread, saw that David was in the same place. With each wave he was pushed to the bottom of the ocean, holding his breath while tumbling underwater like a rag doll. I could not safely help him. The bubbly whitewater was fully aerated, which made swimming and even treading water difficult. David and I both knew the secret to survival was getting out of the impact zone as fast as possible before the next wave. But David wasn’t moving.
In moments that felt like hours, David eventually began to back paddle away from the impact zone. But then I lost him and saw nothing after the next wave. By the end of the set of waves, the ocean, from the beach to where the waves broke, was blanketed with thick white foam. Surfers were scattered along the foamy inside and began to paddle slowly back out to where they could catch waves.
I finally saw David on the beach, sitting quietly, his legs scrunched up to his body. Another surfer had paddled toward David, pulled him on his board, and brought him to the beach. He was fine, but shaken. “I almost drowned,” he said.
I’m not certain David has recovered from that shock, even years later. Until that frightening day he’d only experienced the joys of surfing and had not fully realized the sea’s immense power. Even though he was a strong swimmer and for years a competitive springboard and platform diver with the ability to confront his fears, the terror of trying to swim out of the highly aerated whitewater stayed with him. Six years later he still refused, perhaps wisely, to paddle out on big days at Venice Beach.
Once bitten, twice shy is an apt way of thinking about how and why we learn to fear threatening things. One fearful experience can have a profound influence on our perceptions of safety and well-being, as David’s story so aptly shows, and we see evidence throughout the animal kingdom of individuals learning what to fear and what not to fear. Human fear can manifest as anything from mild anxiety to the symptoms associated with post-traumatic stress disorder. If we wish to understand how individuals learn to fear, we must realize that experience and context influence assessments of risk. We are constantly updating our assessments of risk with additional information.
Modifying behavior based on experience is an adaptive process that increases the number of descendants successful animals leave, which means it’s often location- or context-specific and subject to trade-offs. We can learn based on our own experiences or on others’, and social learning can be an amplifier that permits life-enhancing lessons to spread quickly and widely. Learning is an essential adaptation, one shared by many species. There are many insights to be gained by thinking about how experience can be used to modify behavior, whether in animals or in humans. Understanding how learning influences behavior is becoming more and more critical for planning how we will manage our ever-increasing interactions with wildlife.
This chapter will address a variety of issues associated with learning. I hope to provide you with tools and insights to better understand why we learn what we learn. In many cases, we learn to fear, which means we should be able to unlearn fear too.
Learning can be broadly defined as a process by which experiences change an individual’s behavior over time. It’s important to note that these behavioral changes should not be attributable to either time or developmental processes alone. For instance, we would not say that an animal has learned to fear a predator if an older animal runs faster from a predator just because it’s older, stronger, and has longer legs. Likewise, we would not attribute David’s avoidance of surfing Venice Beach to learning if avoiding water was part of the natural process of growing up for all humans. We would, however, say that learning has taken place if individuals exposed to predators flee faster than they did before exposure or if only those children who had almost drowned now avoid large surf.
Learning to fear requires one or more bad experiences. It may be a rapid process, or it may occur over time. For example, as discussed in Chapter 3, many species of fish are able to learn to associate a particular chemical cue with a predator the first time they encounter both together. This one-trial, rapid learning makes a lot of sense if you think about its function. David’s response to almost drowning wired into his brain a fear of large, powerful waves at Venice Beach. This makes sense from an evolutionary standpoint: fearful stimuli should lead to rapid learning because individuals who learn to fear appropriate things will leave more descendants on average than those who take longer to learn to flee potentially fatal experiences. Natural selection has selected for organisms that do their best given the constraints that they face. A surprising trend has emerged from this process of natural selection: some animals appear to learn in a Bayesian way.
An eighteenth-century minister, philosopher, and statistician, Thomas Bayes, developed a logic of decision making based on accumulating evidence—learning. Differing from traditional statistical logic, Bayesian logic assumes that we have some prior knowledge about the likelihood of an event, and with experience we update our estimate. For example, a traditional statistical approach would assume that there is a fifty-fifty chance of being struck by lightning when venturing outside, which essentially means that being struck by lightning is random. By contrast, Bayesian logic assumes that on a sunny day there is much less than a 50 percent chance of getting struck by lightning, but if you are on a golf course or alpine peak during a thunderstorm, there is a much greater chance. Additional information—for instance, the distance between you and each successive lightning strike—improves the estimate. Bayesian logic learns from prior experience. Formally, one begins with what’s called a prior probability distribution and updates the “prior” based on accumulated evidence. This new posterior probability distribution is, as Bayesians would assert, the best estimate of an event occurring—whether it’s a lightning strike or an attack by an eagle or a terrorist.
Updating risk assessments based on new evidence should be ubiquitous. Some animals have been shown to behave as though they are using Bayesian logic. For these, we assume that natural selection has selected for reasonable priors. This is particularly true when we think about predation risk.
Animals living with snakes, for instance, might be more prone to respond alarmingly to a sudden encounter with a long, thin object than those living without snakes. Or dense, vegetated habitats may be avoided because obstructed visibility increases predation risk. When animals find themselves in novel environments—which occurs with increasing frequency when humans intentionally or unintentionally modify the habitat, or shift animal populations in a way that changes their landscape—we often see suboptimal outcomes that could be based on a now-faulty Bayesian prior. As an example, consider animals recently relocated to snake-free environments. If there are no snakes around, jumping back in response to a slightly curved stick that may look like a snake is an unnecessarily time-consuming response. Or, consider the customary benefits of not foraging near shrubs that may contain a sit-and-wait predator like a snake. If snakes are no longer present, avoiding shrubs unnecessarily eliminates access to potentially valuable food. We expect that, on average, animals making suboptimal foraging decisions will leave fewer descendants. More generally, we expect that natural selection will select for ways to eliminate costly responses if those costly responses are no longer needed.
The rate and conditions under which a species learns things is itself subject to natural selection and is the product of evolution. Tammar wallabies required only four exposures to Andrea Griffin in her witch’s costume to learn to associate the fox with an aversive experience. They may have needed fewer exposures, but we wanted to be sure that they had a sufficient opportunity to learn, so we tested them after four exposures. Remember, tammars had some ability to respond fearfully to foxes without training; tammars were programmed to learn to increase their fearful responses to things that they had some fear of already. In contrast, they didn’t think much of our taxidermic goat, which did not elicit a fearful response, and even four pairings of Andrea with the goat did not teach them to fear it.
Learning can be both rapid and location- or context-specific. The film Rosewater tells the true story of journalist Maziar Bahari, who was imprisoned by the Iranian government. Brutally interrogated over several months of detention while blindfolded, he could only identify his interrogator by his rosewater-enhanced scent. People who have been through experiences such as his often have traumatic flashbacks that are triggered by seemingly innocuous things—like the scent of rosewater—or other specific environmental features that were present during a traumatic event.
For individuals that encounter different sorts of predators, learning is an essential process that permits individuals to respond appropriately to salient stimuli, like an attacking eagle, while not responding to nonthreatening stimuli, like a falling leaf. But animals and humans can become overcome with alarm if they learn to fear the wrong things. For species that suffer only predictable risks, learning is potentially costly. For tammars, learning to fear a nonpredatory goat rather than or in addition to a predatory fox would waste a lot of time and energy.
In cases where risks and cues are predictable, and there is insufficient time to learn, we expect innate predator recognition abilities. A fascinating example of this is seen in Karen Warkentin’s work with accelerated hatching in red-eyed tree frogs. Tree frogs lay their eggs in masses on tree branches, and until they hatch they are extremely vulnerable to snake predation. What’s truly remarkable is that if a snake approaches an egg mass, the reasonably well-developed eggs have the ability to rapidly hatch (often in seconds, and on average much less than a minute), and the prematurely hatched tadpoles fall like rain from the egg mass. Karen has elegantly shown that the embryos react to specific vibrations that would normally be produced by a slithering snake approaching the eggs. Since there is no opportunity for learning from this sort of fatal attack, the embryos respond innately to these vibrational stimuli.
But many species work from some innate predator recognition abilities of scary sights, sounds, and smells, and further hone them with experience. Sometimes, these recognition templates permit the identification of novel species. We assumed that this was what permitted Kangaroo Island tammars to respond to European red foxes—a species with which they had absolutely no experience in their lifetimes and with which they did not evolve.
Some of the best contemporary behavioral ecological research on learning is conducted on fish. My friends and colleagues Maud Ferrari and Doug Chivers have carried out many of these experiments. Maud and Doug—a dynamic husband and wife duo—initially come across as opposites. Maud rapidly fires out an unending stream of incredibly novel and exceptionally clear ideas while Doug offers stellar logic and impressive data sets in a quiet, paced delivery. I always look forward to speaking with them at scientific conferences, where I learn from their both intensive and extensive work studying the nature of fear in aquatic systems.
The basic fish learning experiment involves holding a predatory fish in a tank until its associated chemicals become concentrated in the water. The water from this tank is then piped into a tank holding a prey species. An extract is made by grinding up bits of the prey species (often just the skin, but on occasion the whole animal), mixing it with water, and filtering it to remove small particles. It is then piped into the prey’s holding tank. Boom! As soon as this skin extract, representing deceased prey, diffuses through the water, the prey learn that the predator’s smell indicates danger. They respond fearfully when they detect the predator.
Working with larval damselfish on an island in Australia’s Great Barrier Reef, Doug and Maud and their colleagues asked how the background level of risk influenced the ability to learn about nonthreatening things. As they note, learning about nonthreatening things is as important as learning about threatening things, yet it is studied less often. If learning is Bayesian, then we would expect that there are both cues of safety and cues of risk, but most studies focus on the risk. And if you’re already in a risky environment, then the same sort of risk cue should mean different things about your probability of survival. After all, as discussed before, whether you’re on an alpine peak or inside your home should change how you feel about lightning.
To study the effects of background levels of risk, Doug and Maud pre-exposed larval damselfish living with no risk and those living with some risk of predation (which they simulated by exposing them to predator odors over a series of days) to a novel, nonthreatening odor. The novel odor rapidly lost any salience in the fish exposed to no risk of predation. But the fish exposed to some risk of predation never learned that the novel scent was nonthreatening; they kept up their vigilance even to potentially benign stimuli because they lived in a high-risk environment. Background level of risk thus influences how animals learn what to fear and, conversely, what is safe. Organisms in risky environments are typically well attuned and, if anything, overestimate the risk of novel stimuli until more experience shows them to be benign. We will learn more about the logic underlying these biased assessments in Chapter 10.
Fish can be quite sophisticated in their ability to learn the specific level of risk that their predators create. For example, the concentration of skin extract should indicate the risk that a predator poses. Wise fish should infer that there is a greater risk when there is more skin extract in the water. And they have been shown to do this. Maud and Doug trained fathead minnows to learn that brown trout represented either a high risk or a low risk. They manipulated risk level by changing levels of skin extract, reasoning that more extract would be associated with either a closer predator or more predators. In the high-risk condition, the minnows responded fearfully by dashing around their tank and hiding in response to the brown trout, as well as by reducing their activity once sheltered. Sometimes the fish sought out other fish in the tank to shoal with. In addition to the direct response to the brown trout, the minnows also generalized this fearful response to a close relative of the brown trout—rainbow trout. What is fascinating is that they only made this generalization in the high-risk condition, presumably because they are very sensitive to particular risk cues—even if they are not perfect cues from a member of their own species. Tellingly, they did not generalize this fearful response to a distant relative of brown trout—yellow perch—under either the high-risk or low-risk condition. As species diverge over evolutionary time, they often smell less similar, and this limits the ability for one species to use other species as cues for predation risk.
We’ve found that learning transfers across similar sorts of predators. This is because predators often share similar scents, particularly if they’re eating similar prey. Predators that hunt their prey in similar ways have converged on similar appearances, known as archetypes. Archetypes may be visual, acoustic, or olfactory. In the visual domain, the now-extinct thylacine, or marsupial wolf, with its long mouth, long legs, and wolflike body resembles a wolf or dog despite not being at all related to them because they hunt their prey in similar ways. Thylacines, like wolves, chase their food down (which requires long legs) and then immobilize it with their mouth (which requires a long and toothy snout). Acoustically, closely related predators like coyotes and wolves have similar vocalizations because of their shared ancestry. The archetype hypothesis is a way to show that prey recognize more than a single species and may even respond to novel predators.
Social learning from other members of their own species can act as a potent amplifier of fear. As part of her dissertation work, Andrea Griffin asked whether tammar wallabies could learn from others to become more fearful of foxes. After training a “demonstrator” wallaby to fear a fox (by the aforementioned taxidermic fox, net, and witches’ hat), she paired the demonstrator with a naïve tammar—one who had no prior exposure to foxes. The demonstrator hopped away in alarm when the fox appeared. After only a few experiences, the formerly naïve tammar responded fearfully to the fox also.
But not all socially transmitted fear is adaptive. Humans sometimes experience hysterical contagion, which occurs when people either convince each other that something is the source of fear or they copy the behavior of others. A famous example of this occurred in Salem, Massachusetts, in 1692. It’s thought that the adolescent girls, who began acting strangely and were accused of being witches, as well as the fearful members of Salem themselves represented a case of mass, socially transmitted hysteria. Not all such cases of hysterical contagion are associated with fear. A much more pleasant example is the Tanganyika laughter epidemic of 1962, involving uncontrolled laughing by schoolchildren. Although the children were not laughing constantly, it spread through the schools and may have taken a year to peter out.
In recent decades humans have started to transmit fear via technology. In 1994, during the Rwandan genocide, Hutu extremists used the radio to accelerate the spread of the massacre. They broadcast hate-filled messages and instructions to “kill the cockroaches”—the Tutsis. This social transmission encouraged more than 800,000 deaths in 100 days. In 2014, the Ebola epidemic’s coverage on television and the Internet triggered socially transmitted hysteria. Fears were multiplied by our now twenty-four-hour news cycle, which favors maintaining viewer attention over providing reasoned analysis. For instance, Representative Duncan Hunter (R) from California suggested on the Sean Hannity Show that Islamic State terrorists who were infected with Ebola were attempting to cross into the United States through weak southern borders and infect the population with Ebola. This completely unfounded rumor was immediately refuted by US government officials. However, once unfounded rumors are created, they persist in perpetuity on the Internet for all to reread and spin to achieve their own outcomes.
Besides its potential as a force multiplier of fearful rumors, social transmission leads to desensitization to real or manufactured threats. Further to the American response to the Ebola outbreak in 2014, social transmission of fear exaggerated the likelihood of a breakout of the virus from cases being treated in hospital isolation wards. Or consider the US government’s assertions in 2003, prior to the Iraq War, that Iraq had weapons of mass destruction. Through the mere repetition of incorrect assertions, people came to believe that Iraq had the capability to destroy the planet.
Indeed, there’s evidence that repeating false messages can change people’s perceptions of the truth, even when people have concrete knowledge that the message is incorrect—a phenomenon referred to as illusory truth. Ronald Reagan dubbed the Soviet Union the “Evil Empire” in 1983, George W. Bush referred to Iran, Iraq, and North Korea as the “Axis of Evil” in his 2002 State of the Union address, and in 2018, Donald Trump referred to many immigrants in a caravan of Central American refugees marching toward the US border seeking asylum as “stone-cold criminals.” When such statements are repeated and repeated, people’s perceptions are changed whether or not there is supporting evidence for the claim, and many view these nations or groups more fearfully. Such perceptions have consequences for our foreign policy. As we will discuss in Chapters 10 and 12, there are reasons why we are particularly susceptible to messages that involve gory deaths caused by a virus, or biological or chemical weapons. But we should realize that we are particularly susceptible to messages that invoke fear, whether plausible or not.
If our assessments of risk emerge from Bayesian processes, however, we should be able to eliminate our fears. We’ve known one way that fear can be eliminated for at least 2,500 years.
There once was a shepherd boy who was bored as he sat on the hillside watching the village sheep. To amuse himself he took a great breath and sang out, “Wolf! Wolf! The Wolf is chasing the sheep!”
Aesop, the fifth century BC storyteller, wrote long ago about the process of habituation, though he didn’t use that name. Habituation leads to declined responsiveness to a stimulus, as well as its doppelganger—sensitization. As the fable teaches, the shepherd boy quickly loses the attention of the citizens once they realize his cries of wolf are unfounded. They learn that he is not truthful about the predator. So when a wolf appears, no one believes his calls, and no one comes to his defense.
While intensive research in the last century has led to well-supported generalizations about mechanisms of habituation, we have not yet developed a natural history of habituation. A natural history would help us predict how wildlife will respond to humans and anthropogenic stimuli. The need for predictive models has never been greater because our growing human population is urbanizing and increasingly seeking out encounters with wildlife.
As both urbanization and nature-based tourism expands, animals around the world are being exposed to humans. In 1950 only about 64 percent of Americans lived in urban areas, while in 2018, 82 percent did. In 2018 an estimated 55 percent of humans lived in urban areas, and according to the United Nations, by 2050 68 percent of the global population will do so. This rapid urbanization has had major consequences for the animals and plants that evolved in ecosystems with far fewer people. One strategy to preserve biodiversity is to set aside protected areas, such as parks, reserves, and wilderness areas, where human impacts will be reduced. Yet these protected areas receive an estimated 8 billion visits per year; a number that exceeds the number of people on Earth, and then some. Developing a fundamental understanding of how animals respond to humans is essential if we wish to preserve the life-sustaining biodiversity that surrounds us. As discussed in Chapter 6, animals initially respond to approaching humans as if they are predators, by fleeing and thus diverting time and energy away from other important activities, such as acquiring food, resting, or searching for predators.
Habituation requires repeated unthreatening exposures to a formerly fearful stimulus. To study the process of habituation it’s essential to look at an individual’s responses over time. If an animal habituates to repeated exposures to humans, it will, for instance, tolerate closer approaches before fleeing. Thus, if we find that flight initiation distances in an urban population are much shorter than those in a rural population, we would infer that the urban population is more tolerant of humans than the rural population. Habituation may explain this increased tolerance in urban areas, but other processes could also explain this resilience. For instance, the urban population may be composed entirely of human-tolerant animals. Or animals may sort themselves out according to their degree of tolerance: those individuals that are urban tolerant will be found in urban areas while those that are afraid of people will avoid urban areas. And last, it’s also possible that natural selection has created urban-tolerant animals. Nevertheless, if we wish to understand whether habituation is occurring, we must follow individuals and track their responses over time and increased exposures.
Many urban populations of birds, mammals, and lizards allow humans to get closer to them than rural populations of the same species. We assume that this reflects repeated, benign contact and thus some degree of habituation. But this tolerance, while widespread, is not ubiquitous. Sometimes animals react in the opposite manner. Developing a natural history of habituation would help us better understand when animals are likely to become more sensitive to human presence.
Sensitization may be adaptive if it helps animals avoid potentially risky or costly situations. Elephants, remarkably, get quite upset when they hear bees buzzing around. Their response makes a lot of sense; elephants may avoid getting their sensitive trunks stung. Humans who are allergic to bees might react in a similar way. We are likely sensitized to sensational claims about Ebola, whether true or false, because we want to avoid getting horrific diseases.
But we should not always assume that sensitization is adaptive. For instance, drug addiction, in humans as well as in animals that serve as experimental models for drug addiction, involves sensitized responses. Experience with drugs like cocaine or methamphetamines increases the desire for more drugs. Worse for addicts is that the underlying neural circuits and neurochemistry involved in sensitization share many components with the underlying neural circuits associated with drug-seeking behavior. This suggests that the process of sensitization may not be adaptive in this situation but rather may prime individuals to seek out more drugs.
I used these insights about human visitors and our methods of studying flight initiation distance (FID) described in Chapter 6 to work with birds in Southern California. The habitats I studied differed based on the number of human visitors. In the course of my study, I noticed that out of fourteen species of California coastal chaparral birds studied, only four species had significantly different FIDs in those areas with relatively more human visitation than areas that were less frequented by humans. However, and somewhat unexpectedly, these four species fled us at greater distances when exposed to more people. The other ten species had no significant differences in FID as a function of quantifiable differences in human visitation. What could explain this pattern of apparent sensitization to humans?
It’s important to note that this finding was in stark contrast to my studies of wetland birds in Southern California. The California coastal wetlands have mostly been filled in and turned into homes and businesses. The few wetlands that persist play a vital role in providing resting stops for migrating birds and wintering grounds for the snowbirds that have left the arctic winter for warmer, sunnier climes. On these wetlands, all of the species we studied were more tolerant of human approaches when they were routinely exposed to more humans, consistent with habituation.
I suggested that perhaps species living in limited habitats, such as the remnant wetland fragments in Southern California, may be more likely to habituate than those living in more contiguous habitats, such as the coastal chaparral, because those that live in remnant fragments may have already gone through a filtering process that eliminated less-tolerant species or individuals. The filtering process drove local extinctions of those species that simply could not tolerate humans. Thus, the only species that occupy these wetlands are those that are, to some degree, tolerant of humans. I called this idea the “contiguous habitat hypothesis.” My hypothesis needs proper testing by evaluating it in other habitats, ones with more species and different types of disturbance. We still don’t really know what precisely it is about humans that disturbs animals. Is it the number of encounters with pedestrians, people walking dogs, or vehicles? If so, is this a simple dose-response relationship whereby more people are more disturbing or is there a threshold that, once exceeded, causes disproportionately large impacts? Are they bothered by the smells or sounds or light associated with us? We know that all of these stimuli can have negative impacts on animals, but we need to develop a much better understanding of how, precisely, humans influence wildlife.
So we know that differences in historical exposure to predators as well as current exposure to predators may allow us to predict the degree to which a species will be prepared to learn about fearful things. And we suspect that having no options to leave a habitat fosters the ability to habituate, while having options to move off may foster the ability to sensitize to recurrent disturbances. As discussed in Chapter 6, large species are more likely to be disturbed by people, but large species that can coexist with humans are also more likely to become tolerant of them with repeated, benign exposures. Body size is an important life-history trait that is associated with other life-history traits. The rules animals have evolved that enable them to respond adaptively to potential threats are the result of past selection. The degree of match between current and historical threats should permit us to understand how and when species may (or may not) habituate to specific types of disturbance. Truly novel disturbances may be more difficult to habituate to than those that share features with other known threats. These generalizations mark the start of a more predictive understanding that can explain the conditions under which animals, and perhaps humans, learn to fear or learn to forget fearful stimuli.
Theo, our corgi, loved to bark at people passing through a parking lot next to our home. While squat, corgis are not really that small, and his bark was impressive and loud. Theo had spent time in Colorado, where afternoon thundershowers are common in July and August, but there is almost never thunder and lightning on the west side of Los Angeles, where we live. One day, while Theo was barking at people from our patio table, a thunderstorm struck our neighborhood with little warning. With the first clap of thunder Theo ran into the house, barking with fear and closely following Janice. Thunder is a very low-frequency sound, and it’s loud, of course. I suspect Theo thought there was a much bigger animal outside, and he wisely feared it. We ended up wrapping his torso in a scarf to calm him down. Ever since this event he startles, alert with fear, when we’re watching a movie that includes a sudden low-frequency rumble.
The fact that a single fear-inducing event can have lasting impacts on our sense of security, and seemingly Theo’s, is most strikingly seen in post-traumatic stress disorder (PTSD). Exposure therapy, which can reduce PTSD symptoms, relies on desensitizing people to a formerly fear-inducing stimulus. But there may also be pharmacological therapies.
To understand how exposure therapy works and to develop potential drugs to treat PTSD, we need a model system. Much of the work on PTSD has used fear-conditioning studies in rats. When electric shocks are paired with a specific stimulus, rats quickly learn to avoid that stimulus. But PTSD is also characterized by nonspecific anxiety. My UCLA colleague Michael Fanslow has developed this anxiety in rats by shocking them without providing any predictive trigger. These anxious rats will then learn, with only a single trial, to respond fearfully when a stimulus is later provided. Studies have documented neurological changes in their fear-conditioning circuits that resemble the neurological changes associated with human PTSD. This creates a set of animals with which he and his colleagues are able to research pharmacological treatments to eliminate the disorder. Recent work in his lab has shown that by blocking stress hormone receptors in the brain, rats fail to become fear conditioned.
If this finding can be applied to humans, it may create therapies that reduce PTSD. I would certainly have considered taking something after my Kenyan attack to avoid permanently changing how I assess risks. Janice and I would have discussed giving David something to reduce his trauma after nearly drowning on Venice Beach. And for anyone who has been violently assaulted, such medicines may prevent years of trauma.
For those who already suffer from PTSD, the aforementioned exposure therapy is used to attempt to eliminate it. Recent work with rats has identified the possible neural basis of why it takes specific exposure to fear-inducing stimuli to erase fearful memories. Neurons in the dentate gyrus, located in the temporal lobe, are associated with both memory formation and memory extinction. Therapeutically, the idea is that by desensitizing people to the fear-inducing triggers, they will no longer suffer from the debilitating fears and paralysis that often characterize the disorder. If, for instance, someone’s PTSD was triggered by an attack they experienced in a vehicle, they may have learned to associate being in a car with an increased likelihood of an attack. Therapists work with patients to repeatedly create safe experiences of being in cars. With enough support from their therapists and lots of practice, this prolonged exposure therapy may eliminate the fear-inducing PTSD triggers.
Fearful responses have been honed by natural selection to ensure our safety. We share an evolved ability to form traumatic memories from traumatic experiences, which for many species is essential for survival. Fish learn quickly about predatory threats, and tammar wallabies know they should fear predators that look like foxes, but not those that look like herbivorous goats. But exposure to new threats—dangers that our ancestors have not experienced—creates new challenges.
We all are prepared to respond adaptively to new challenges because we (as well as many of our ancestors) are able to modify our behavior based on experiences. We, and other species, can learn from each other, and this acts to increase the rate at which knowledge can spread through a population. But the lessons from life show us that it’s not simply adaptive in many situations to learn; we must properly match our assessments of risk with reality. Learning is Bayesian. At a more proximate level, desensitization and habituation provide mechanisms to reverse traumatic events. And insights from these mechanisms provide the potential tools to enhance our coexistence with wildlife in an increasingly urban world.