The Pleasure Instinct: Why We Crave Adventure, Chocolate, Pheromones, and Music - Gene Wallenstein (2008)
Part I. The Pleasure Instinct and Brain Development
Chapter 3. What Makes Sammy Dance?
There seems to be a continuing realization by psychologists that perhaps the white rat cannot reveal everything there is to know about behavior.
—Keller and Marian Breland, The Misbehavior of Organisms
The mind of the thoroughly well-informed man is a dreadful thing. It is like a bric-à-brac shop, all monsters and dust, with everything priced above its proper value.
—Oscar Wilde, The Picture of Dorian Gray
One morning in 1970 a tortured twenty-four-year-old man with a history of drug abuse and severe depression walked into Dr. Robert Heath’s office at Tulane Medical School in New Orleans. By then Heath was a well-known, albeit controversial, figure who founded the Department of Psychiatry and Neurology at Tulane in 1948 after being recruited from Columbia University. Within a year of joining the faculty, Heath and his coworkers were conducting experimental studies in humans that would forever change the way psychiatrists think about emotions and at the same time provide enough source material to keep biomedical ethicists busy for decades to come.
By the time he was twenty-four years old, the patient known as B-19 had a diagnosis of temporal lobe epilepsy compounded by a history of chronic drug abuse and depression.“I live with the idea of suicide daily,” he is quoted as saying, and it is reported that he made several “abortive attempts.”We also learn that B-19 was homosexual and that “one aspect for the total treatment program for this patient was to explore the possibility of altering his sexual orientation through electrical stimulation of pleasure sites of the brain.”
During the early years of his tenure, Heath pioneered the therapeutic use of electrical stimulation of the brain (ESB) to treat epilepsy. Impressed by the work of Olds and Milner, who had just discovered “pleasure centers” in the brains of rats, Heath adapted their approach to recondition the brains of patients suffering from affective disorders and particularly schizophrenia.“The primary symptom of schizophrenia isn’t hallucinations or delusions,” he told a reporter years later.“It’s a defect in the pleasure response. Schizophrenics have a predominance of painful emotions. They function in an almost continuous state of fear or rage, fight or flight, because they don’t have the pleasure to neutralize it.” The idea was tantalizing—just stimulate the neural pleasure centers of a schizophrenic and this might rekindle damaged circuits affected by the disease and enable the patient to once again experience positive emotions.
Electrodes and cannulas (needle-thin tubes through which drugs may be delivered directly into the brain) were placed in fourteen subcortical structures of B-19’s brain, including the septal region, hippocampus, amygdala, and hypothalamus—areas that were hypothesized to regulate emotions in humans and had previously been identified as locations where rats “self-stimulate.”
Prior to the study, B-19’s “interests, contacts, and fantasies were exclusively homosexual; heterosexual activities were repugnant to him.”After B-19 recovered from the surgery, Heath and his coworkers stimulated each electrode briefly and asked their patient to report what he felt. Stimulation at most brain regions produced only mild or “neutral” feelings, and in some cases actually induced anxiety or other aversive sensations. But one electrode positioned in the septal region consistently produced an intense pleasurable response. “The patient reported feelings of pleasure, alertness, and warmth (goodwill); he had feelings of sexual arousal and described a compulsion to masturbate.”
During the first phase of treatment, B-19 was given a portable transistorized device that could be used to activate the different electrodes implanted in his brain.At first he experimented by stimulating a variety of sites—each time he pressed a different button, the device sent out a one-second pulse of electrical current to the corresponding electrode. Within a short time, however, the young patient was stimulating his septal electrode almost exclusively. During treatment sessions, he was permitted to wear the device for periods of three hours, and on one occasion stimulated this region more than fifteen hundred times (about once every thirteen seconds on average). During phase two of the treatment, B-19 was allowed to self-stimulate his septal electrode while watching “stag movies” of heterosexual activity, and he became “increasingly aroused.” Pleased with their patient’s progress, the innovative scientists hired a “lady of the evening” to assist them with phase three in which B-19 had his first “pleasurable” heterosexual encounter after being primed by five minutes of continuous septal stimulation.
ESB was used to treat hundreds of patients (not just at Tulane) through the 1970s, although with limited success in schizophrenics. An interesting observation was that patients suffering from depression or anxiety often rated septal stimulation as more pleasurable than patients without a mood disorder. At the time, it was believed that the stimulation was restoring the functioning of a weakened limbic system—a set of brain regions that regulate emotional valence in humans; however, this interpretation has been refined considerably in recent years. Neuroscientists now understand far more about the limbic system and how it communicates with neocortical structures during the expression and feeling of emotions. Most of what we know about the biology of pleasure began with an accidental discovery by two young scientists.
The Nature of “Natural” Reward
In many areas of science rapid advancement often comes from serendipitous discoveries. Working in a basement laboratory in 1954, newly doctored James Olds and graduate student Peter Milner were smoothing out the kinks of a study in which they were to implant electrodes deep into the reticular formation of rats. The German physiologist Rudolph Hess had recently shown that stimulation of the brain-stem regulates the sleep-wake cycle, and Olds believed that different sites within this region might selectively lead to either activation or inhibition of the neocortex, producing states of alertness or calm respectively.
During the first run of their experiment, each time the rat sniffed a particular corner of the square testing platform, Olds stimulated its brain, expecting that the activation would initiate the animal’s natural tendency to explore and visit other corners. Strangely, just the opposite happened—the rat returned again and again to the corner where it received the stimulation. Puzzled by this, the pair soon confirmed that the electrode had not been positioned correctly in the reticular formation, as thought, but rather landed in the septal region, a largely unexplored, phylogenetically ancient part of the brain.
Realizing they were on to something important, they quickly replicated their findings and developed another experiment where each rat was allowed to directly self-stimulate its septal region by pressing a lever in a testing chamber—a twist on the classical Skinner box, where rats learn to press a lever for access to food or water. To their surprise, this produced rapid learning of the lever press response, and their rats were willing to perform a variety of tasks to have access to this stimulation. In other words, a brief electrical pulse to the septum seemed to have very similar reinforcing properties to natural rewards such as food, water, and sex. In the fifty years since this original study, self-stimulation has been found to reinforce behavior at a number of distinct but related brain regions in the limbic system and across a variety of different species, ranging from goldfish to humans. The fact that this neural circuit (and its purported function) is conserved across such diverse species suggests that it is a phylogenetically older part of the brain, having evolved in an ancestor common to all of these groups.
Perhaps no other discovery in neuroscience has created such a torrent of experiments, conferences, publications, and additional questions. As the circuit continued to be charted by the early pioneers it became apparent that brain stimulation was not only rewarding, it was also drive-inducing, and thus became a tool for studying natural motivation. Yet the big question remained unanswered:What exactly does an animal experience when its septum is stimulated? Is it pleasure? Is it sexual in nature? Or is it a general state of arousal that amplifies the natural drives of an animal depending on the contextual cues that surround it? Clearly we can’t ask a rat for commentary, so we have to infer its inner state—whether we’re talking about motivation, drives, feelings, or some other operational term—from its behavior. As we shall see, this is never easy.
All animals, large and small, slow and fast, have mechanisms that allow them to adapt to changes in their local environment. Even single-celled organisms use chemotaxis as a means of guiding themselves along chemical gradients toward nutrient-rich environments. We typically call these motivated behaviors in that they refer to any adjustments, internal or external, made by an organism in response to environmental changes. Often, these adjustments are regulatory, designed to maintain homeostasis, and they can include modifications to endocrine, autonomic, immune, or behavioral processes.
When Olds and Milner made their original discovery, motivation was largely thought to be a simple matter of drive or “need” reduction. This theoretical perspective works fine if we limit our discussion to thermoregulatory or ingestive functions—for instance, perspiring to dissipate body heat, or thirst to satisfy the need for water. However, it fails to explain other behaviors such as aggression, sex, or novelty-seeking, all of which can be triggered by an external stimulus, yet have no identifiable deficit state. It also fails to explain why in many cases normal homeostatic mechanisms can be overridden by strong external incentives, such as occurs during drug binges, while gorging ourselves on a delicious meal, or flying down a snow-capped mountain on two thin slabs of fiberglass. Instead, we typically explain these sorts of behaviors as an attraction to external stimuli or events that have appetitive or rewarding properties.
Unfortunately for most animals, food is not just splayed out for the taking; sexual partners are not lined up and waiting; and there is not always natural spring water nearby. All animals have to actively seek out these sources. Thus, motivated behavior is not simply eating a meal or engaging in sex; these consummatory responses are typically only the end points of a long and complex sequence of actions, guided in part by drives and in part by the appetitive features of an incentive. So if appetitive features of an environment—things that are inherently attractive or rewarding to animals—are not simply tied to essential needs or drives that ensure survival, how do they emerge? Are they innate or learned?
The short answer is both. Behavioral scientists have used an impressive variety of strategies for studying motivation, many of which take advantage of the fact that animals exhibit both classically conditioned (Pavlovian) reflexes and goal-directed instrumental (also called operant) behaviors. An often-used operant conditioning paradigm involves placing a hungry rat in a test chamber and making the delivery of food contingent upon some behavior. Say, for example, that whenever a small lever is pressed by the rat, a food pellet is automatically dispensed into the testing chamber. The rat, of course, has no innate or implicit knowledge of this relationship—it has to discover it by exploring the environment.
When placed in a novel situation, mammals typically exhibit a period of freezing behavior, where they stay in one place and examine their surroundings followed by a gradual increase in exploratory activity. During the exploratory phase, a rat will eventually press the lever, often accidentally, and, after a few co-occurrences of lever press- food appearance, gradually discover that a relationship exists between the two. This process is known as associative learning, and occurs in virtually all animals, from sea slugs to primates. It is the foundation on which most forms of learning are based. In our example, learning the link between the lever being pressed and the subsequent appearance of food is facilitated by the fact that food, in this case, is a positive reinforcer, meaning its appearance increases the likelihood of repeating the behavior that preceded it.
Understanding how associative learning works has preoccupied the minds of psychologists, philosophers, and biologists for decades. Of particular importance to the present discussion is that not all associations are learned with the same accuracy and speed. In general, associative learning occurs most easily when one of the components of the pair involves an evolutionarily important variable. For instance, rats usually learn the association between a specific food item and the sickness it induces after only a single exposure, and they use this learning to avoid these foods in the future. Such conditioned taste aversion has obvious benefits to the survival of an organism, allowing it to avoid ingesting dangerous and potentially lethal toxins. The same is true of fear. Rats often learn to avoid places where they have experienced foot shock after just one trial.
Contrasting this, rats typically need hundreds of trials to learn an arbitrary association between two neutral stimuli, for example an odor and a small object, if neither of the objects belongs to a broader class of stimuli that have had a significant impact on the evolution of the species. Indeed, some associations seem impossible to learn if the components are in opposition to species-specific tendencies. In their classic (and mirthful) paper “The Misbehavior of Organisms,” Keller and Marion Breland reviewed a series of failed attempts at using operant conditioning techniques to teach animals to perform simple tasks.The title was a playful jab at their teacher, B. F. Skinner, whose book The Behavior of Organisms is widely considered a seminal work in the field of behavioral analysis.
In their first example they asked, “What makes Sammy dance?” Sammy, it turns out, is one of many adult bantam chickens that have been trained to emerge from a holding compartment and stand on a platform for twelve to fifteen seconds, after which food is automatically dispensed. In this task the only requirement for reinforcement is that each chicken must depress the platform and wait. Simple enough; however, most of the chickens developed a pronounced tendency to scratch at the platform, a behavior that became even more persistent when the waiting period was lengthened. Although the Brelands could not train chickens to perform the original task,“we were able to change our plans so as to make use of the scratch pattern, and the result was the ‘dancing chicken’ exhibit.” The point of this article was not to demonstrate that capable trainers can outsmart poultry, but rather that after being conditioned to perform a specific response, animals can gradually drift into entirely different behaviors that seem to go directly against reinforcement contingencies. “It can clearly be seen that these particular behaviors to which the animals drift are clear-cut examples of instinctive behaviors having to do with the natural food-getting behaviors of the particular species. The dancing chicken is exhibiting the gallinaceous birds’ scratch pattern that in nature often preceded ingestion.” Reinforcing the chickens with food thus led to the emergence of innate behavior that anticipated, but was not in itself rewarded by, the arrival of food.
Some stimuli, as noted before, have supernormal features, meaning their reinforcing value cannot be accounted for simply in terms of drive reduction. For instance, although bland foods such as normal rat chow do not reinforce the behavior of a satiated rat, sweet foods high in natural sugars do. In fact, sugars rated most sweet-tasting by humans, such as sucrose and fructose, are stronger reinforcers of behavior in satiated rats than those ranking lower on the sweetness scale, such as lactose and glucose. Sugar is obviously an evolutionarily powerful stimulus. Knowing how to detect it in the environment and having a taste preference for it so that it would be ingested provided clear survival benefits for our hunter-gatherer ancestors who possessed these traits.We now realize the biological importance of sugar molecules as a precursor source of ATP, which powers so many of the biochemical reactions critical for cell functioning. Our ancient brethren certainly had no understanding of these processes—they needn’t have for survival. For individuals to gain a selective advantage, they simply needed a hedonic preference for foods that contained digestible sugars and a means for their detection and ingestion.
Hedonic preference refers to a stimulus property that is innately reinforcing or, speaking more colloquially, naturally rewarding. Psychologists often use the terms primary positive reinforcer or unconditioned reinforcer to describe this type of stimulus, emphasizing the notion that their rewarding properties are usually in place at or before birth. Human newborns, for example, prefer sweet-tasting liquids rather than plain water immediately after birth, before ever being exposed to sugar in the external environment (a preference that facilitates breast-feeding, since mothers’ milk is high in lactose).Two other primary positive reinforcers that work through taste include saltiness and a newly discovered gustatory dimension called umami, an indicator of protein content that is produced by monosodium glutamate (MSG). Both can serve as primary positive reinforcers of behavior and are discussed in the chapter on the development of taste preferences (see chapter 6).
There are primary positive reinforcers in every sensory domain—touch, smell, taste, vision, and hearing. As we shall see, many behaviors such as kin identification, parent-offspring attachment, and some forms of communication are motivated by compound reinforcers—particularly attractive combinations of primary positive reinforcers from several sensory modalities.
Another rich source of incentives that motivate complex behavior depends on conditioned positive reinforcers (also known as higher-order reinforcers)—initially neutral stimuli or behaviors that become rewarding through an association with a primary positive reinforcer. For instance, if our hungry rats learn that lever-pressing brings food, then this behavior itself acquires incentive value.1 After learning the association, rats press the lever constantly, often to the exclusion of other behaviors such as exploration and grooming. However, they obviously have no innate fondness for this behavior, since before learning they seldom exhibit it, and only then at random. Clearly the incentive value of lever-pressing is contingent on the subsequent appearance of food, because this behavioral tendency terminates once food is no longer available—a phenomenon known as extinction.
Hedonic preferences, in conjunction with the thoughts, perceptions, and actions that become conditioned to them, provide fertile ground for studying pleasure and other emotions. Yet, it’s important to realize that this theoretical foundation does not stem from a behaviorist school of thought, in which we are born as blank slates waiting to be writ upon. Quite the contrary, it assumes there are evolutionary and developmental constraints that shape what we are likely to learn, when we are likely to learn it, and how such learning takes place. Before we discover the ways in which each sensory modality contributes to the hedonic palate, however, we will first examine how this ancient circuitry evolved in our species and consider how the process parallels the embryological development of the human nervous system.
Since soft tissue does not fossilize, our understanding of human brain evolution comes mainly from comparative studies of living species that are closely related to Homo sapiens.There have been discoveries of early primate skulls that happened to fossilize with an imprint of the former owner’s brain, allowing an endocast to be made of the cortical surface; however, artifacts such as these tell us nothing about the inner circuitry of these ancient brains and very little about their gross anatomy. Instead, scientists have used a different approach.
The relationships among the major vertebrate classes have long ago been organized by examining the anatomical characteristics that distinguish species.This system of classification has been refined through the years by making further comparisons based on fossilized bone remains and DNA fragments in an effort to reconstruct the phylogenetic history of living forms. Powerful new methods have been used recently to compare DNA fragments of living animals from different branches of the phylogenetic tree and have found that in many cases (but not all) the phylogenetic reconstructions based on genetic evidence correspond well with those based on studies of fossilized remains.
These reconstructions can be used to determine whether similarities between two animals are the result of a shared evolutionary history or rather co-evolved independently in both species. For example, since chimpanzees, apes, and humans evolved from a common ancestor some four million years ago, it is likely that the brain structures common to all of these species were present in that ancestor. Brain regions that are not common to all three species are most likely to be evolutionarily newer structures derived from earlier predecessors. Similarly, brain areas that are common to all Old World monkeys (for example, Macaca) and hominoids (for example, apes and humans) are most likely even older in that they stem from the common ancestor class of Old World anthropoids that existed more than twelve million years ago.
Understanding approximately when (in terms of evolution) a brain region came into existence helps us in several ways. First, if we can show that other anatomical changes co-appeared at the same time as the brain region in question—for instance, the development of bipedalism or forward-facing eyes—it may be possible to reconstruct the selection factors that contributed to its evolution. Second, since we have a growing understanding of the relationship between brain and behavior, information about the general evolutionary history of the human brain allows us to develop theories about the evolutionary history of our cognitive and emotional capacities.
The evolution of mammals from reptilelike ancestors about three hundred million years ago brought the gradual accumulation of several distinctive features: the appearance of hair; sweat glands; mammary glands and suckling behavior; specialized teeth for grinding, slicing, and piercing new food sources; and physiological mechanisms for maintaining a constant body temperature (thermoregulation). All of these adaptations suggest that early mammals led a predatory lifestyle. During this period, the brain also went through a number of significant transformations.
Chief among these is the development of the neocortex, the vast outer portion of the cerebrum that constitutes about 85 percent of the human brain’s total mass and that is responsible for higher-level cognitive functions such as language, learning, memory, and complex thought. Reptiles and birds possess an anatomically simpler, three-layered version that is sometimes referred to as archicortex (old cortex). Mammals have a six-layered version that is far denser in cell counts per volume and contains a greater diversity of cell types; however, they also possess several three-layered cortical structures, such as the olfactory cortex and hippocampus (collectively referred to as allocortex), which are similar to the archicortex of nonmammals. For these reasons, it is believed by many (but certainly not all) neuroanatomists that the neocortex is unique to mammals and derived from phylogenetically older cortical areas.
This evolutionary ordering of three-layered cortical areas predating the six-layered neocortex is paralleled in the embryological development of humans. By the sixth week of gestation, a human fetus will already have many brain-stem structures partially developed. These will control basic physiological functioning in the newborn—respiration, sleep-wake cycles, thermoregulation, and a host of motivated behaviors. Brain-stem sites are followed by the initial appearance of subcortical regions such as the thalamus and hypothalamus sometime around the tenth week. As we shall see, many of these brain regions play a pivotal role in the generation of motivated behaviors and act as sensory integration sites. By fourteen weeks the allocortical areas start to develop, eventually becoming part of the limbic system, a region responsible for learning, memory, and processing emotional information. By the sixteenth week, cells begin to appear in what will eventually become the neocortex. However, it’s important to note that none of these areas is wired together functionally yet.This occurs in two distinct stages—neurogenesis and synaptogenesis.
How the Developing Brain Gets Wired
Our understanding of how the human brain develops from a smooth sheet of ectoderm2 into the mature adult form has changed radically in just the past ten years. Scientists now have a much deeper appreciation for the way developing brains depend on specific types of stimulation patterns and sensory experiences to activate important genes. These insights have come from studies of developing infants using noninvasive brain imaging techniques, neuropsychological experiments, and an improved understanding of how genes actually work.
Genes have two basic components, a template (or coding) region that provides information about how to make a specific protein, and a regulatory region that determines when a gene is expressed or repressed. The template region is what we usually think of when we hear the phrase “it’s in our genes.” The information in this region is not modified by experience or learning, only through mutations, which are rare and essentially random. The regulatory region, on the other hand—the on/off switch of the gene—is highly sensitive to a host of experiential factors.
A great variety of signaling proteins can bind to the regulatory region of a gene and modulate its subsequent transcription and expression. These signaling proteins are known as transcription regulators. Put simply, when transcription regulators bind to a segment of a gene, they activate its expression and the eventual production of a new protein. Thus they have direct control over whether a gene is turned on or off.
A number of factors influence the way a transcription regulator binds to a gene. Both internal and external stimuli (that is, things we experience) activate signaling pathways that result in alterations of this binding process. Some signaling pathways are differentially activated as a result of the normal developmental process. Others are activated by stress, learning, hormonal changes, or social/experiential interactions.
For example, psychological and physical stress causes the release of the adrenal gland steroid glucocorticoid (also known as cortisol), which circulates in the peripheral and central nervous system (brain and spinal cord). In the brain, this steroid activates a transcription regulator that binds to the regulatory region of several genes, inducing the transcription and expression of new proteins involved in the long-term regulation of the stress response. Hence, social factors such as stress regulate gene expression and the subsequent production of specific proteins. Indeed, all forms of learning are incorporated into our biological makeup in the altered expression of specific genes that encode the production of selective proteins in brain cells.
Gene expression can be extremely selective in targeting the production of proteins unique to a specific type of nerve cell and brain region. A particular experience, say a psychological stressor, will result in the production of a very specific set of proteins, while another experience, for example, learning a new phone number, will result in a different set. These experience-induced changes in gene expression and subsequent protein production are not transmitted from generation to generation genetically. None of these alterations in gene expression is incorporated into the sperm or egg, and therefore are not heritable.All changes in gene expression that result from learning or being exposed to experiential/environmental factors are transmitted culturally rather than genetically, and they clearly have a profound impact on the way brains develop.
The fact that genes have essentially two functional components has important implications for development and the relationship between nature and nurture. Being familiar with the way genes really work makes it easier to see why most biologists have long ago given up on the nature versus nurture debate as a false dichotomy. The genes that code the way brains are built do not contribute to development unless they are transcribed and expressed. Hence, experience is an essential part of development even at the level of the gene.
How do genes build brains? As I write this chapter my wife is four and a half months pregnant, and every day brings new questions about little Kai’s development. The typical adult human brain has about 100 billion nerve cells or neurons. Each neuron connects to thousands of others, resulting in about 1014 (a 1 followed by 15 zeroes) different connections. How, then, do the 25,000 or so genes identified by the Human Genome Project code such a combinatorically large and complicated system? Clearly, since the numerical differences are so great, genetic information does not uniquely specify where every single neuron resides or where each of its thousands of connections will terminate. Instead, genes specify more general rules for neuron development and migration.
At the earliest stages of development, we start off as three primitive cell layers: endoderm, which consists of cells that eventually line our internal organs and vessels; mesoderm, destined to become the major structural components of the body, including bones and muscle groups; and finally, ectoderm, which becomes the central nervous system, skin, hair, and nails. Kai’s entire mental world—his thoughts, emotions, sensations, and perceptions—emerge from this thin sheet of cells. Within the first few days of gestation, the primitive cell layers elongate and fold into a cylindrical tube called the notochord, the progenitor of Kai’s spinal column. This process recapitulates the earliest event in the evolutionary transition from invertebrate to vertebrate forms that occurred more than 600 million years ago.
Once the notochord is formed, it guides the ectoderm layer, which progresses through a series of well-defined stages, first thickening and then folding in on itself to form the neural tube. At about nineteen days into gestation, just about the time when Melissa and I first learn she is pregnant, the earliest form of Kai’s future brain and spinal cord begin to emerge through a process called neurogenesis. During this period, the front end of Kai’s neural tube develops three enlargements, which eventually become the two cerebral hemispheres and the brain-stem. The neural tube then goes through a rapid growth spurt, where the entire cycle from cell division to cell division takes place in about an hour and a half. Some of these precursor cells are destined to become neurons, while others will mature into glial cells, which serve a variety of supportive functions in the brain.
As cell division and replication continue, the three enlargements begin to take on more detail, eventually forming all the major components of Kai’s brain. At two months into gestation he is little more than two inches long, yet all of his major brain structures have begun to take shape, including the elementary forms of the medulla, pons, and midbrain, which combine to form the brain-stem; subcortical structures such as the thalamus, hypothalamus, and basal ganglia; then a bit more slowly, the allocortex; and even more slowly, the neocortical regions. Kai’s brain will develop from the bottom to the top, with lower brain-stem structures such as the medulla maturing first, followed in sequence by the upper brain-stem, subcortical areas, the allocortical regions, and then the neocortex.
It’s often asked why neurogenesis starts at the bottom of the brain and progresses toward the top (or since the brain is three-dimensional, from the inside of the brain to outer regions). A clue can be found if we compare the embryological development of an individual with the evolutionary development of our species. Until roughly four weeks of gestation, the embryo that will become Kai is practically indistinguishable from embryos of many bird, reptilian, and mammalian species. But by the sixth week, he begins to look more and more like a mammal, and by week seven he appears decidedly primate. As a general rule, species of similar phylogenetic forms tend to look alike for longer periods during development. For example, human and chimpanzee embryos share strikingly similar features until about the seventh week of gestation, at which time they begin to diverge in appearance. Human and rabbit embryos, on the other hand, share similar developmental features only until about the fourth week of gestation. Thus, embryological development is highly conserved across different species—those structures appearing earliest in development are most common across species and thought to be derived from phylogenetically earlier forms.
This makes sense when one considers the vast diversity of adult life forms that mature from a single fertilized egg. Evolution is driven by the natural selection of forms that are altered through mutations, and chances are much greater that an offspring will be viable if a mutation occurs at a developmentally later rather than earlier stage. Mutations that occur early in embryological growth tend to have devastating consequences because they alter all subsequent stages of development. Consequently, successful adaptations are most often “added on” or modified from existing structures that were present in earlier phylogenetic forms.
This relationship between ontogeny and phylogeny explains why human brains take so long to develop and why certain regions mature before others. The human brain, particularly certain areas of the neocortex, distinguishes us from other primates and takes significantly longer to develop and mature than other systems, such as those that control respiration and circulation. First to come online are those brain regions that are involved in the essential behaviors common to most mammals—for example, simple reflexive movements and those that eventually control respiration and digestion. Many of these simpler behaviors depend on neurons in the spinal cord or lower brain-stem.
By the end of the first trimester, Kai has a fairly mature brain-stem and diencephalon, which consists of the thalamus and hypothalamus. The thalamus is a critical sensory relay station that takes part in a two-way dialogue with different areas of the neocortex, sending new information through and often being instructed by the neocortex in a top-down manner to filter certain features of sensory information so that it never reaches conscious perception. The hypothalamus is just as important because it will serve as the mediator between the rest of Kai’s brain and his endocrine and immune systems. His first sensations of hunger and satiety, and the emotions that accompany these survival behaviors, will be mediated largely by cells in his newly formed hypothalamus.
Even within a specific sensory modality—sight, for example—those features of vision that are most common across species, such as detecting movement and contrasting light intensity, arise functionally well before other features, such as color vision and depth perception, which are more unique to primates.As mentioned above, this is because the brain regions that control each of these sensory functions arise in a specific sequence laid down by our genetic programming that reflects a gradient from functions common to many species to functions more unique across phylogenetic classes. We start off broad functionally, and with development continue to add those features that are first common to all mammals, then those that are only seen in primates, and finally those that only humans possess.
After he is born, Kai will be able to see things that move long before he’ll be able to see colors. This is because he will emerge from the womb with only brain-stem, thalamus, and the most primitive visual cortical regions connected and working. And as will become clear, these circuits are in a very sensitive state. For the processes of neurogenesis and synaptogenesis to continue the job of connecting his entire visual brain and fine-tuning it, the circuits he has thus far must be stimulated by specific types of visual experience. It is in the service of this developmental necessity that pleasure—driven by the hypothalamus and other limbic structures that come online early—participates to ensure that newborns seek out the forms of sensory experience that optimally stimulate the time-sensitive growth and maturation of their brains.
This can be thought of as a “bootstrapping” mechanism, whereby developmentally early brain systems that support essential survival and regulatory processes contain the functions that ensure the further development of higher-order brain systems involved in more complicated forms of perception and cognition. Just as Kai’s brain will need nutritionally rich sources of metabolic energy to continue normal development, so too will it need to encounter sensory-rich stimulation at specific intervals during maturation. The experience of pleasure thus provides newborns with a general rule for seeking out patterns of stimulation that guide the normal development of all higher cognitive functions, including perception, language, and abstract reasoning—functions that have traditionally been treated as being entirely separate from their more passionate brethren.
At four and a half months into our adventure, Melissa—a prepregnancy vegetarian—can now keep her food down, and occasionally sends me out (often very late!) to pick up a Whopper with extra pickles. It is at about this time that the processes of neurogenesis and cell proliferation begin to peak in Kai’s brain.The numbers are staggering. To reach the estimated one hundred billion neurons that comprise a human brain—the majority of which are in place by midgestation—Kai will have to produce an average of five hundred thousand cells per minute during the first four and a half months. During this time of cell proliferation, neurons migrate to their intended locations, and once there begin to grow extensions toward other cells in a process called synaptogenesis.
One hundred billion neurons alone do not make a brain. It’s in the detailed wiring that connects cells together within and between different regions that we find the mechanisms of sight, smell, hearing, touch, and the multitude of unique capacities that make us human. Each neuron has three basic parts—the cell body, which contains most of the metabolic machinery; dendrites, which take in information from other cells; and an axon, which carries information from the cell to target cells with which it will communicate. After neurogenesis, brain cells extend their dendrites and axons, and form synaptic connections with other cells. A synapse is essentially a point of communication between two neurons, and the dialogue is electrochemical. So important is this process to life that its absence, as measured by a lack of significant electrical activity in scalp EEG recordings, is taken as a definition of death.
It’s been estimated that between midgestation and two years of postnatal life—the peak period of synaptogenesis—close to 15,000 synapses are produced on every cortical neuron. This averages to an astounding 1.8 million new synapses formed every second during this period. At present, there are several controversial theories as to how a cell finds its correct target cells. For example, how do the output cells of the retina know they must bypass certain brain regions and terminate their axons in the proper portions of the visual thalamus? And how then do those thalamic cells know to project to specific regions of the primary visual cortex? Although no single theory is consistent with all of the available data, it is widely accepted among neurobiologists that synaptogenesis is a comparatively long process—continuing through gestation and several years of postnatal life—and works primarily through competition.
Nature begins the job.The general wiring scheme is laid down by our genetic programming and follows a very specific developmental sequence of time-dependent growth patterns that resembles that observed during neurogenesis. The first areas to undergo significant synaptogenesis are in the lower brain-stem, followed by the upper brain-stem, the diencephalon, various subcortical regions, the allocortex, and finally neocortical regions. Neuroscientists are only beginning to understand how genes code the basic wiring of the brain, and the ways in which this process depends on experience. Genes, it turns out, direct axons and dendrites only to their approximate target locations. Once these beginner circuits start to function, however, experience and the pleasure instinct take over to fine-tune the connections and shape them into a precise network unique to a child’s environment.
From the end of gestation through early childhood, Kai will produce about twice as many synaptic connections as he will eventually need in his adult brain. Cells make and receive thousands of synapses during this promiscuous period, resulting in a large-scale but rather diffuse communication system between different brain regions. The overproduction of synapses is an evolutionary trade-off—since the small number of genes we possess simply cannot dictate each of the quadrillion or so connections between brain cells, they instead provide general rules for making contacts between areas. Once these diffuse connections are in place, nature takes over and begins to selectively prune certain synapses based on the types of stimulation patterns (that is, experiences) the brain receives. In this way, experience, guided by the pleasure instinct, refines the communication network within the brain, making certain connections stronger while removing others.
Synaptic pruning is a requirement of normal brain development. As I mentioned, it works directly through a process of competition—survival of the fittest. Each of the synaptic connections in Kai’s brain has the potential to survive past this competitive period of pruning, but about half will perish, and along with them certain functional capabilities.The rule is simple—use it or lose it.To survive, the synapse between two cells must be activated consistently.Those synapses that are activated the most have three advantages over less active synapses. First, when activated, they tend to inhibit surrounding synapses (through cellular responses) that are competing for stimulation. Second, the electrochemical dialogue between active cells triggers a series of biochemical reactions that strengthen the synapse, cementing it in place. And third, synapses that are inhibited during this competitive period have an increased likelihood of triggering active processes that weaken and eventually terminate the connection. These fundamental discoveries from neuroscience have profound implications for the way we raise our children and educate our young during the first two decades of life.
While the major work of neurogenesis and synaptogenesis is finished relatively early during development, synaptic pruning occurs very slowly, lasting at least until an individual’s early twenties and perhaps longer. Although it takes decades, synaptic pruning generally follows the same sequence as neurogenesis and synaptogenesis, progressing from lower structures of the brain-stem that tend to be more broadly represented across phylogenetic class, upward toward structures such as the prefrontal cortex that are most unique to primates. One consequence of this sequential development is that the behavioral, perceptual, and cognitive functions that depend on each of these regions also tend to arise through a specific—and culturally independent—sequence.
Once a particular brain region undergoes synaptogenesis and the overproduction of synaptic connections, this marks the onset of abilities regulated by that region, such as the emergence of color vision, sound localization, and language. Although experience influences all stages of development, it is predominantly the long period of synaptic pruning that fixes the overall quality and nuances of these abilities. As synaptic pruning refines certain abilities, for example, a capacity to discern phrase boundaries unique to human languages, the process also results in the sacrifice of alternative synaptic connections that gradually make it difficult to acquire other abilities, such as the perception of sounds that are not normally a part of human experience. Hence, synaptic pruning is essentially a selection process driven by experience.
As we shall see, the behaviors that emerge during synaptic pruning—guided by the pleasure instinct—are really forms of self-stimulation that infants, toddlers, children, and adolescents must produce to ensure normal brain development given the ecological and environmental context in which that development and growth take place. A central theme of this book is that nature has solved this functional requirement in primates (and perhaps other mammals) by exploiting the early capacities of limbic structures to produce pleasurable sensations, which have been associated through our long evolutionary lineage with optimal forms of brain stimulation.
From culture to culture, it is striking that infants and toddlers pass through the same developmental milestones, many of which can be used to gauge how far their brain has matured. Once a potential capacity arises, they must self-stimulate in ways that encourage the further development of the brain systems that mediate these and related behaviors. We find this process at work in every sensory modality, and the sequence of self-stimulating behaviors that emerge reflects the order in which our primary sensory systems come online during development.As we will learn in the chapters that follow, this developmental progression is an echo of our evolutionary past.