Scienceblind

1 | WHY WE GET THE WORLD WRONG

MOST PEOPLE TODAY WOULDN'T CLASSIFY MILK AS A HEALTH HAZARD. To us, it's an innocuous form of nutrition, poured over cereal or consumed with cookies. Milk was not always so innocuous, though. Only a century ago, it was a leading cause of food-borne illness in the industrialized world. Drinking cow's milk is not inherently dangerous—humans have been doing so for millennia—but it becomes dangerous if too much time has passed between when the milk is collected and when the milk is consumed. Milk is typically consumed without heating, and heat is what kills the bacteria inherent in our food. Milk is also high in sugar and fat, which makes it a perfect medium for bacterial growth. The negligible amount of bacteria present in milk when it is collected grows exponentially with each passing hour—a biological fact that milk consumers never really grappled with until the dawn of the Industrial Revolution, in the latter half of the nineteenth century.

The industrial revolution changed the landscape of where people worked and thus where they lived. As the population of Europe and the United States shifted from the country (where people worked on farms) to the city (where they worked in factories), people no longer lived close to the cows that produced their milk. Dairy farmers began transporting milk farther and farther from its source, which meant that people began drinking milk longer and longer after it had been collected. This combination of factors—that milk is consumed cold, that milk is an ideal incubator for bacteria, and that milk was being consumed days after its collection—led to several mass epidemics of milk-borne disease in Europe and the United States; the epidemics included outbreaks of tuberculosis, typhoid, scarlet fever, and cowpox. Milk in the nineteenth century was, according to one medical expert, "as deadly as Socrates' hemlock."

The problem of how to safely consume milk several hours (or days) after its collection was solved in the 1860s with a relatively simple process: heating milk for long enough to kill most of the bacteria within but not so long as to alter its sensory qualities or nutritional value. This method of food treatment, devised by Louis Pasteur, came to be known as pasteurization. The health consequences of pasteurization were immediate and immense. Those most at risk of milk-borne disease in the nineteenth century were infants, as infants fed cow's milk were several times more likely to die than those fed from the breast. After pasteurization was introduced, however, infant mortality rates in urban centers dropped by around 20 percent.

Today, pasteurized milk is considered one of the safest foods to consume, associated with less than 1 percent of all food-borne illnesses. Oddly, however, people are increasingly opting to drink unpasteurized milk, and as a consequence, the rates of milk-borne disease are rising. Between 2007 and 2009, the United States experienced thirty outbreaks related to the bacteria Campylobacter, Salmonella, and E. coli—outbreaks linked to the consumption of unpasteurized milk. Between 2010 and 2012, that number rose to fifty-one. People are increasingly buying raw, unpasteurized milk for a variety of reasons: the belief that raw milk tastes better than pasteurized milk, that raw milk has higher nutritional content than pasteurized milk (which it does not), that raw milk is what humans were intended to drink, and that consumers should have the right to choose whether their milk is pasteurized or not. Those who reject pasteurized milk in favor of a more "natural" alternative do so seemingly blind to the fact that, before the advent of pasteurization, thousands of people suffered organ failure, miscarriage, blindness, paralysis, and even death at the hands of milk-borne diseases.

Do people fully understand what they are rejecting when they reject pasteurizing? Probably not. Pasteurization is counterintuitive. It's counterintuitive because germs are counterintuitive. Germs are living things that cannot be seen; they are passed from one host to another without detection, and they make us ill several hours or days after we come in contact with them. Also counterintuitive is the idea that germs transform our food from sources of nutrition to sources of disease but that we can stop germs from doing so by killing them with heat. Heating food to kill germs is a widespread practice in the food industry. Several types of food are pasteurized (or otherwise heat-treated) before they hit the shelves—beer, wine, juice, canned fruits, canned vegetables—not just milk. It is ironic that advocates for unpasteurized milk are seemingly okay with pasteurized beer and canned peaches. Either they believe that forgoing pasteurization is a justified risk for milk but not for other foods, or, more likely, they fail to understand what pasteurization is and why it is a necessary safeguard against food-borne illness.

The science behind pasteurization is as sound as science gets, but many people reject this science. They reject not just the science behind pasteurization but science in general, from immunology to geology to genetics. A recent survey of American adults found that only 65 percent believe that humans have evolved over time, compared with 98 percent of the members of the world's largest scientific society, the American Association for the Advancement of Science (AAAS). Only 50 percent of American adults believe that climate change is due mostly to human activity, compared with 87 percent of AAAS members. And only 37 percent of American adults believe that genetically modified foods are safe to eat, compared with 88 percent of AAAS members.

Science denial is not a modern phenomenon. Most people denied that the earth orbits the sun, that the continents are drifting, or that disease is caused by germs when those ideas were first proposed. Still, resistance to science in today's age—an age flush with scientific information and science education—requires explanation. Many scholars and media pundits point to ideology as an explanation, either political ideology or religious ideology. Others point to misinformation, as when vaccines were falsely linked to autism or when genetically modified foods were falsely linked to cancer. All these factors have been shown to play a role in science denial. Conservatives are less likely to accept science than liberals, religious individuals are less likely to accept science than secular individuals, and misinformation breeds skepticism and hostility toward scientific ideas. But these factors are not the only causes of science denial. Psychologists have uncovered another: intuitive theories.

Intuitive theories are our untutored explanations for how the world works. They are our best guess as to why we observe the events we do and how we can intervene in those events to change them. Intuitive theories cover all manner of phenomena—from gravity to geology to illness to adaptation—and they operate from infancy to senescence. The problem is, they are often wrong. Our intuitive theory of illness, for instance, is grounded in behavior (what we should and should not do to stay healthy), not microbes. Thus it seems incredible that heating milk could render it safer to drink or that injecting dead viruses into our bodies, as done in vaccination, could confer immunity to live strains of the disease. Likewise, our intuitive theory of geology assumes that the earth is a static object, not a dynamic system, and thus we find it inconceivable that humans could be changing the earth itself, causing earthquakes though hydraulic fracking or global warming though carbon emission.

Intuitive theories are a double-edged sword. On one hand, they broaden our perspective of the phenomena they seek to explain and refine our interactions with those phenomena because holding an intuitive theory is better than holding no theory at all. On the other hand, they close our minds to ideas and observations that are inconsistent with those theories, and they keep us from discovering the true nature of how things work. They can cause us not just to misconceive reality but to be blind to reality—to ignore facts and findings that definitively refute those theories. Thus, my goal in writing this book is to introduce you, the reader, to your own intuitive theories and to help you appreciate when and how those theories may lead you astray.

In this book, I hope to convince you of two main ideas. The first is that we do, in fact, get the world wrong—that our intuitive theories in several domains of knowledge carve up the world into entities and processes that do not actually exist. The second is that, to get the world right, we need to do more than just change our beliefs; we need to change the very concepts that articulate those beliefs. That is, to get the world right, we cannot simply refine our intuitive theories; we must dismantle them and rebuild them from their foundations. Galileo once decreed that "all truths are easy to understand once they are discovered; the point is to discover them," but he was wrong. There are many, many truths that are not easy to understand, because they defy our earliest-developing and most easily accessed ideas about how the world works. This book is a story of those truths: why they initially elude us and how we can come to grasp them.

THE IDEA THAT we construct intuitive theories strikes many, at first, as an overly intellectualized view of how we think about the world. Why would a nonphysicist construct a theory of motion or a theory of matter? Why would a nonbiologist construct a theory of inheritance or a theory of evolution? We do so because physics and biology are inescapable aspects of human life; we are immersed in physical and biological phenomena every day.

We may not care about motion in the abstract, but we care about lifting boxes and pouring cereal, riding bicycles, and throwing balls. We may not care about matter in the abstract, but we care about melting ice and boiling water, preventing rust, and lighting fires. Likewise, we care about inheritance insofar as we want to know how likely it is we'll go bald or whether we are predisposed to cancer, and we care about evolution insofar as we want to know why bacteria develop drug resistance or where dogs came from. While few of us can articulate a detailed theory of matter that can explain both rusting and burning or a detailed theory of evolution that can explain both drug resistance and canine domestication, we have coherent and systematic ideas about those phenomena nonetheless.

Psychologists call intuitive theories intuitive because these ideas are our first attempt to understand the phenomena around us, before we learn scientific theories of those same phenomena. They call intuitive theories theories because these ideas embody a specific kind of knowledge: causal knowledge. Causal knowledge is an understanding of cause-and-effect relationships. It allows us to make inferences from our observations—inferences about why something happened in the past (explanation) or what is likely to happen in the future (prediction).

Much of the causal knowledge embodied by our intuitive theories is learned through experience, but at least some of that knowledge is innate. Whether a piece of knowledge is learned or innate is an empirical question—a question that psychologists answer by studying people of various ages and experiences. Studies with young infants, for instance, suggest that many of our expectations about motion and matter are innate. Studies with adults from different cultures, on the other hand, suggest that many of our expectations about illness and cosmology are shaped by what we hear from the people around us. Nevertheless, all of our intuitive theories are shaped both by innate expectations and by lived experiences. Infants may come into the world with expectations about the behavior of physical objects, but those expectation are further refined by their experience interacting with objects. Likewise, different cultures may subscribe to different theories of illness, but all such theories are grounded in the shared experience of what illness actually looks like (e.g., coughing, congestion, fevers).

Intuitive theories vary not just in their source but also in their assumptions about causality. Most theories posit causal mechanisms of a natural (i.e., ordinary) flavor, but some posit mechanisms of a supernatural flavor. Causal mechanisms of a natural flavor are, in principle, observable and controllable. They are often labeled with scientific terminology—for example, heat, inertia, gene, natural selection—but they do not actually correspond to scientific ideas. What scientists mean by a term like heat (energy transfer at the molecular level) is a far cry from what nonscientists mean by the same term (an immaterial substance that flows in and out of objects and can be trapped or contained). Causal mechanisms of a supernatural flavor, on the other hand, are beyond the observation and control of mere mortals. Those mechanisms have no counterpart in science—for example, karma, witchcraft, souls, God—but they still provide systematic explanations for natural phenomena (e.g., displeased ancestors) and systematic means of responding to those phenomena (e.g., sacrificial offerings). And supernatural explanations are often no less substantive than natural ones. Karma, for instance, is no less substantive an explanation for why we get sick than "cold weather" or "bad air," and divine creation is no less substantive an explanation for where species come from than "transmutation" (a sudden change in form) or "spontaneous generation."

Given that we live in a thoroughly scientific world, you might wonder whether intuitive theories are a dying breed—something we constructed in the past, for want of scientific information, but we will stop constructing in the future as scientific information becomes more available and more accessible. Rest assured, intuitive theories are not a dying breed. They are a permanent fixture of human cognition because they are the handiwork of children, and children are not likely to be affected by changes in the availability or accessibility of scientific information. It's not because children have shorter attention spans than adults have or because children are less interested in the natural world than adults are. It's because children lack the concepts needed to encode the scientific information we might teach them.

Take, for instance, the concept of heat. Children can sense an object's warmth—or how efficiently the object transfers heat to them or from them—but they cannot actually sense its heat, as humans have no sensory apparatus for registering the collective motion of a system's molecules. To understand the scientific concept of heat, children must learn a molecular theory of matter. We do, of course, teach children a molecular theory of matter but not until they have reached middle school, and by that time, they have already constructed an intuitive theory of heat—a theory that treats heat as a substance rather than as a process (discussed in Chapter 3). We could attempt to forestall this event by introducing a molecular theory of matter earlier in children's education, but a molecular theory is itself counterintuitive. How do you explain a molecule to a preschooler, let alone an electron or a chemical bond? And how do you forestall a child from mapping thermal language—heat, hot, cold, cool—to concepts that the child already understands, which, in this case, are concepts like substance, containment, and flow?

Clearly, many of us do learn a scientific concept of heat, but the task is not trivial. It requires devising an entirely new framework for thinking about thermal phenomena—a framework that differs in kind from the one we create on our own. Psychologists call this type of learning conceptual change. This is not your run-of-the-mill learning, like learning the traits of an unfamiliar animal or the history of an unfamiliar country. Psychologists call that kind of learning knowledge enrichment. What differentiates conceptual change from knowledge enrichment is whether, at the outset of learning, we possess concepts capable of making sense of the information we need to learn.

Knowledge enrichment is the process of using old concepts to acquire new beliefs, as when we use the concepts of whales, breathing, and air to acquire the belief that whales breathe air. Conceptual change, on the other hand, is the process of acquiring new concepts or, really, new types of concepts. If I told you that a species of mouse in the Amazon eats humans, I will have invited you to entertain a new concept—the Amazonian man-eating mouse—but that concept is just a subtype of other concepts you already know (mouse, which is a subtype of animal, which is a subtype of living thing). We have no trouble learning new instances of preexisting types; that's just knowledge enrichment in disguise. The trouble comes in learning new types.

An analogy with Legos is useful here. A basic Lego set consists of strictly rectangular blocks. With enough rectangular blocks, you can build anything, from a life-size giraffe to a life-size statue of Conan O'Brian (both of which have been built). But there are certain structures that you cannot build: cars with wheels that roll, planes with propellers that turn, cranes with hooks that lift. To build such structures, you need to supplement your supply of rectangular blocks with new, specialty pieces: wheels, axles, gears, and crankshafts. Any vehicle you try to build without those pieces will be ineffective or incomplete. We can approximate a car, but we can't actually build a car; wheels and axles are a necessity.

Much like building an operational Lego car, building a scientific understanding of the world requires resources that are unavailable to the novice learner—that is, to a child. Those resources are concepts like electricity, density, velocity, planet, organ, virus, and common ancestor. Concepts are quite literally the building blocks of thought, and like building blocks, they have particular structures and functions. Entertaining the thought "humans share a common ancestor with daffodils" is not possible without the concept common ancestor, and entertaining the thought "water is denser than ice" is not possible without the concept density. Those concepts are not part of our innate knowledge; nor are they learned from everyday experience with the physical world. They require conceptual change.

Conceptual change is a rare, hard-won achievement. It is difficult to initiate and difficult to complete, as I will try to make clear from specific instances of conceptual change presented in each chapter. At this point, however, I hope it is becoming apparent that intuitive theories and conceptual change are intrinsically linked. We construct intuitive theories of natural phenomena because constructing scientific theories of those phenomena requires conceptual change. But to achieve conceptual change, we must overhaul intuitive theories constructed in the absence of a scientific theory. Why we get the world wrong (intuitive theories) is our answer to how to get the world right (conceptual change), but getting the world right cannot happen without first getting the world wrong. It's a circular notion but not hopelessly circular. We are, after all, capable of getting the world right.

INTUITIVE THEORIES ARE a prime source of misconceptions, but they are not the only source of misconceptions. Most of our misconceptions are simple factual errors—typos of the mind. Many people believe that we use only 10 percent of our brain and that the taste buds on our tongue are divided into distinct sections, but neither belief is true. These misconceptions do not signify a profound confusion about brains or tongues; they are just by-products of misinformation.

Distinguishing factual errors from deep-seated misconceptions is critical if we hope to identify (and study) intuitive theories. Many psychologists have grappled with this issue and have come to identify three hallmarks that set intuitive theories apart from other sources of misconceptions. First, intuitive theories are coherent; they embody a logically consistent set of beliefs and expectations. Second, intuitive theories are widespread; they are shared by people of different ages, cultures, and historical periods. Third, intuitive theories are robust; they are resistant to change in the face of counterevidence or counterinstruction.

To get a better sense of these hallmarks, consider the following two thought experiments, which are designed to pump your intuition about physical motion. First, imagine you're standing in a large, open field holding a gun. You aim the gun at the horizon and shoot a bullet parallel to the ground. As you pull the trigger, you drop a second bullet from the same height as the gun. Which bullet hits the ground first, the one you shot or the one you dropped? Second, imagine you're in the crow's nest of a ship sailing at full speed across the open sea. Next to you is a cannonball. You drop the cannonball out of the crow's nest and watch it fall. Where does it land, on the deck of the ship or in the water behind the ship?

If you're like most people, you predicted that the dropped bullet would hit the ground before the shot bullet, reasoning that the shot bullet had been imparted a forward-propelling force that would keep it aloft longer. You also predicted that the cannonball would fall behind the ship, reasoning that the ship would sail out from underneath the cannonball as it fell straight down. Neither prediction is correct, however.

The shot bullet would have no extra force keeping it aloft. As soon as both bullets were released, they would be subject to a single force—gravity—and gravity would bring them to the ground at the same time, albeit hundreds of yards apart. As for the cannonball, it would hit the deck of the ship directly below the crow's nest because the ball would have the same horizontal velocity as the ship on which it was carried. It's true that the ship would sail out from beneath the cannonball's release point, but the cannonball would not fall straight down. It would trace a parabolic path, produced by the combination of its horizontal velocity and its downward acceleration due to gravity, in the same direction as the ship.

Most people's predictions in these two situations are wrong, but that's not because there's something weird about the situations; both involve nothing more than falling objects. Our predictions are wrong because they arise from an intuitive theory of motion that assumes that objects move if, and only if, they have been imparted an internal "force," or impetus. The term "force" has been put in quotations here, because our intuitive notion of force is not what physicists mean by force (the product of mass and acceleration). Forces may change an object's motion, but they are not properties of objects. They are interactions between objects (as discussed in Chapter 5).

That said, our nonscientific beliefs about force and the relation between force and motion are highly coherent. Take, for instance, the two misconceptions primed above: the misconception that an object with horizontal motion (a shot bullet) will succumb to gravity less quickly than will an object with no such motion (a dropped bullet) and the misconception that a carried object (a cannonball) does not inherit the horizontal motion of its carrier (a ship). These misconceptions may seem unrelated, but they are products of the same underlying belief: that projectiles, and only projectiles, have forces imparted to them. We attribute a forward-propelling force to the shot bullet but not to the dropped bullet and not to the cannonball (which was also dropped). The force we attribute to the shot bullet is thought to keep it aloft for longer than the dropped bullet, whereas the absence of such a force is thought to cause the cannonball to fall straight down.

These ideas, though wrong, are internally consistent. They are also incredibly widespread. Impetus-based misconceptions have been found in students of all ages, from preschoolers to college undergraduates. They have been revealed in China, Israel, Mexico, Turkey, Ukraine, the Philippines, and the United States. And they have been revealed even in students who have taken multiple years of college-level physics. You can earn a bachelor's degree in physics and still be an impetus theorist at heart.

This consistency across individuals extends backward in time as well. People have always been impetus theorists, including professional physicists of centuries past. Galileo, for instance, explained projectile motion as follows: "The body moves upward, provided the impressed motive force is greater than the resisting weight. But since that force is continually weakened, it will finally become so diminished that it will no longer overcome the weight of the body." This explanation smacks of impetus, not inertia, and it is the same kind of explanation most of us would provide today, four centuries later. No one today would use the phrase "impressed motive force," but we would express those same ideas with terms like "internal energy," "force of motion," or "momentum." To a physicist, momentum is the product of mass and velocity, but to a nonphysicist, momentum is simply impetus.

What is remarkable about the sustained popularity of an intuitive theory like impetus theory, from Galileo's time to today, is that we've always had reason to doubt it. Impetus theory makes predictions that are never confirmed, because objects move in ways that do not actually accord with those predictions. When a cannonball is shot from a cannon, it traces a fully parabolic path; it never drops straight down as impetus theory predicts it should (because impetus theory predicts that the cannonball will eventually lose its impetus and succumb to gravity). Yet, if we are asked to draw the trajectory of a cannonball shot from a cannon, we draw paths that begin parabolic but end straight down—paths we have never observed, or ever could observe, in real life. Impetus theory can successfully explain some aspects of reality, but it renders us blind to the aspects it cannot.

Impetus theory is not unique in this regard. All intuitive theories are coherent (in their internal logic), widespread (across people), and robust (in the face of counterevidence), and this trifecta gives them a surprising amount of resilience. While we can learn new, more accurate theories of a phenomenon, we can't seem to unlearn our intuitive theories. They continue to lurk in the recesses of our minds long after we have abandoned them as our preferred theory. Intuitive theories are always there, influencing our thoughts and behaviors in subtle yet appreciable ways.