Psychology's Ghosts

Yale UNIVERSITY PRESS

Contents

Chapter One

* * *

Although the public's understanding of science is dominated by images of elegant machines and useful products, the two most basic rituals are making observations and inventing concepts that might explain the evidence. In 1910, Peyton Rous injected cells taken from a rare tumor of the skin found in one chicken into healthy chickens and discovered that the animals receiving the cells developed the same cancerous tumor. After proving that the toxic agent in the tumor cells had to be extremely small, he suggested that a virus was the likely cause of the cancer. In time, many investigators and physicians extended this explanation to all cancers. This belief turned out to be incorrect. We now know that although a virus was the cause of the cancer in Rous's chickens, viruses are not the cause of all cancers.

Every observation is made on one kind of object or process in a specific setting, usually with the help of a particular machine or procedure. Thus there is always the possibility that the same phenomenon might not be observed if the object, setting, machine, or procedure were altered. The scientist's continual hope is that the concept and explanation applied to the observations in one setting with one procedure would remain appropriate in other settings with similar procedures. If so, the conceptual name chosen for the original observations would remain valid. If not, perhaps the particular thing, setting, or procedure was essential to what was observed. Put differently, scientists want to know what exists in nature, but they must always be concerned with the reasons why they believe what they do because different procedures or settings can yield observations that change what they know.

Physicists have been the most successful at imagining concepts and explanations that apply across an extraordinary range of situations, because the major features that define the concepts salt crystal, carbon atom, and electron are relatively stable, inherent characteristics of those objects rather than their functions. Newton's famous equation stating that the gravitational force between two objects is lawfully related to their masses and the distance between them holds for the relation between the earth and moon as well as between an apple on a tree and the earth below. Any object thrown up in the air from any place on the earth will fall to the ground. Einstein's equation E=mc² (energy is the product of the mass of an object and the square of the speed of light) is a quintessential example of the physicist's search for concepts and laws that are indifferent to variations in the setting. This equation does not specify the form the energy assumes (light, heat, radiation) or the object (a kilogram of uranium or iron).

Biologists have had a more mixed success, because the defining features of a large number of biological concepts apply to a restricted number of species. More important, the features of many concepts refer to the reactions of a cell or living form, and the reactions observed depend on the local setting. The presence of one X and one Y chromosome is a defining feature of the concept biological male in mammals, but this feature is not the criterion for maleness in fish or birds. Female fertility is reduced during the winter months among women residing in northern latitudes characteristic of Scandinavia, where there are long periods of darkness, but is not seen among women living close to the equator.

Psychologists have encountered the greatest difficulty generating concepts that applied to settings that were very different from the one that gave rise to the original observation because most concepts refer to the behavioral or physiological reactions of animals and humans to specific situations, rather than to the relatively stable, inherent features of these agents, such as their body mass, pigmentation, or blood type. As a result, many psychological concepts refer to phenomena that are necessarily influenced by the context. The surface of the sea, compared with the sea bottom, provides an analogy. The form that the surface water assumes in any particular ten-thousand-square-foot area is influenced by the wind, the presence of ships of different sizes, whales, and in some places acres of garbage. Yet psychologists continue to search for laws that have the power of E=mc². This practice ignores the fact that a number of physical concepts are mathematical inventions that have not yet been observed and, therefore, lie far below the surface. Dark matter and the Higgs particle are two examples.

A frustrating consequence of the influence of the local context is that few psychological concepts intended to represent a person's tendency to react in a certain way apply across diverse settings, even though concepts, such as fear, extroversion, and intelligence, imply that they do. Most seven-month-old infants cry if an unfamiliar adult with a neutral facial expression walks toward them quickly. But few cry if the same stranger is smiling while walking toward the infant slowly. Therefore, attributing a fear of strangers to an infant depends on the context.

This fact and many others are the source of serious disagreements over the defining features of important concepts. A hand supplies a clarifying metaphor. Biologists study the stable features of the genes, proteins, and other molecules responsible for the development of a hand. Although all hands are by and large similar, the things that hands do, which psychologists measure, vary with the setting.

Physicists agree that quarks, leptons, and bosons, the constituents of atoms, are the basic elements of matter, and they describe the inherent properties of these particles. Biologists concur that genes are the foundation of all the proteins that make up tissues and organs, and they describe genes as sequences of DNA molecules. But psychologists are not even close to an accord on the biological and psychological processes that are the foundations of the phenomena they wish to understand. These include aggressive actions, emotions, consciousness, regulation, morality, stress, and reward. None of these concepts specifies the setting in which the defining information was observed or the procedure that produced the evidence. Therefore, they both imply a generality across contexts and procedures that does not always occur and generate disagreements about the defining properties of these and other popular concepts.

One investigator, for example, will attribute a state of fear to individuals who show a large surge of blood flow to a particular brain site (indicating activation of that site) when a face with a fearful expression appears on a screen. A second scientist will attribute fear to adults who display a large skin conductance reaction on the fingers when they see a stimulus that warns them they might receive an electric shock to the wrist. And a third will attribute fear to college students who report on a questionnaire that they often feel scared when they see a mouse. The problem is that the majority of individuals who display a surge of blood flow to the amygdala to fearful faces do not show a large skin conductance reaction to the threat of an electric shock, and a majority of individuals in the latter group do not say they are afraid of mice. This evidence implies that the concept fear has different meanings in these three settings. It would seem wise, therefore, for investigators to stop writing about "fear" in the abstract and specify the measure and setting whenever they attribute fear to a person or to an animal.

The theoretically important concept reward provides an illustration of the need to include the details of the setting when the concept is invoked. The term reward was invented originally to explain why some events had the power to increase the occurrence of a behavioral reaction that was followed by a rewarding event. The probability that a hungry rat will repeatedly strike a lever with a paw increases if food is delivered after the lever is struck. This fact is interpreted to mean that food is a reward for a hungry animal. Humans, however, find the mastery of a difficult task and the active pursuit of a goal as rewarding. The many writers, composers, and scientists who labor for years in the hope of winning recognition illustrate the wisdom of this view. The feelings and thoughts that follow the sating of hunger, however, are different from the feelings and thoughts accompanying the receipt of praise for an accomplishment. For that reason, the concept reward cannot refer to a single biological or psychological process. Even though one restricted set of neurons is activated by the anticipation of any desirable experience, the total pattern of brain activity is unlikely to be identical to all types of rewards. A small number of neurons in the auditory cortex are excited by music, speech, thunder, explosions, and rain striking a windowpane. But the complete pattern of brain activity, as well as the profile of associations, is quite different to each of these five events, and they therefore should not be regarded as belonging to the same psychological category.

Consider a simple example of the mistaken inference that is possible when it is based on observations in only one context. Let's assume that Mary has a strong motive to please others but a weak need to prove she is well disciplined, whereas Alice has the opposite pair of motives. Both are tested for their ability to inhibit impulsive decisions in a laboratory setting, and both women attain the same score on a measure of impulsivity. But because the identical scores are the result of different motives, the inference that Mary and Alice are equally capable of inhibiting impulsive acts in the laboratory would probably not be affirmed if the setting were a party at which friends urged each woman to stay up all night drinking beer and smoking pot.

There is a more formal way to describe this issue. A majority of psychologists begin their research with a favorite hunch and gather one type of evidence in one setting to confirm the truth of their hypothesis. They are unlikely to try to disprove the favored hypothesis by gathering additional data in different settings. These scientists assume that their hypothesis is true if the evidence they gather supports it. A second, smaller group starts with a weaker intuition about what to observe and begins by gathering a large set of observations in the hope of finding patterns that might generate a strong hypothesis. These investigators assume that what they observe will be the best source of a useful hunch. These are different strategies with different probabilities of generating true statements. The probability that Peyton Rous was correct when he wrote that a virus was the cause of the chicken's skin tumor turned out to be high. But the probability attached to the truth value of the hypothesis that all cancers have a viral origin turned out to be low.

The aims of this chapter can be stated in a few sentences. Behaviors (including verbal replies in humans) and biological responses in a setting remain a primary source of evidence for psychologists who conceive of their task as inventing concepts that will predict when a particular class of behavior will occur and to explain why it occurred. Three factors influence the probability that a particular behavior will be expressed. The properties of the brain and the individual's prior experiences are the most obvious. The structure of the human brain makes it likely that most two-year-olds will smile, speak, and reach for objects in many settings. Life histories add a host of new behaviors that would not have occurred naturally, such as sewing a button, hitting a baseball, or texting a message.

The local setting, the third influence, selects one behavior from an envelope that usually contains more than one possibility. Pedestrians holding an empty candy wrapper can throw it on the sidewalk or put it an official receptacle for rubbish. They are more likely to choose the former if they see others litter but apt to select the latter if they note that others put used wrappers in receptacles. The brain structures of modern humans, which emerged between 100,000 and 150,000 years ago, awarded our species the ability to invent a written script. But the evidence suggests that this ability was not expressed until about 8,000 years ago, when the setting was optimal for the exploitation of this competence. This talent, along with texting a message and operating an automobile, lay dormant for more than 100,000 years.

The local setting affects parents' and teachers' descriptions of the personalities of children. Parents and teachers usually disagree because the parents base their judgments on the child's behavior at home, whereas the teachers rely on behavior in the school setting. Hundreds of psychologists across the United States are involved in intervention programs designed to alter the intellectual abilities or personality traits of children from economically disadvantaged families. In most cases, the intervention occurs in a school, university, or Head Start program. Evaluations of the effectiveness of these efforts reveal that the modest changes observed in the settings where the intervention occurred did not always generalize to the home and community contexts where the parents and children spend most of their time.

Each person is a member of a number of symbolic categories that usually include their developmental stage, sex, family pedigree, ethnicity, religion, class, culture, and work role. Some add their city, nation, or region in which they live. Each of these categories has a privileged link to a particular set of behaviors. The strength of a person's motives, beliefs, and behaviors varies with the categories that are predominant in a particular setting. Because each setting activates only some of these categories, the motives, moods, attitudes, or behaviors that are likely to emerge vary across settings and cultures. The person's categories for sex, age, status, and work role will be activated in male college students who, having volunteered to participate in a study in a university laboratory, meet an older woman who administers some questionnaires. A different pattern of motives and behavior biases will be activated when these same students are at home talking with an older sister. Hence, the answers to a question asking about their feelings toward their parents are apt to differ when posed by a stranger in a laboratory or a sibling at home.

Stanley Milgram attained sudden fame in the late 1970s after publishing a series of studies demonstrating that ordinary Americans would obey an experimenter who told them to administer extremely painful electric shocks to a stranger who they believed was participating in an experiment on learning whenever the stranger made an error. Actually, the stranger was a confederate and was not receiving any shocks. Although a majority of subjects obeyed the examiner and administered what they thought were painful shocks, the features of the setting affected their behavior. The subjects were most likely to administer very painful shocks if the stranger was located in a separate room, the subject could hear his simulated cries of pain, and the experimenter was perceived as a legitimate authority. The subjects were less likely to administer the painful shocks when the stranger was sitting next to them and the experimenter was either physically absent (he gave orders on the telephone) or was not perceived as an authority figure.

The direction of a career is occasionally determined by the distinctive features of a city or nation during a particular era. Claude Lévi-Strauss was France's most celebrated twentieth-century anthropologist, even though he had no interest in this discipline as a young man. Lévi-Strauss had accepted a temporary position as lecturer in sociology in a university in São Paulo because he needed a job. While in Brazil he took a short trip west where he made contact with some Indian tribes, gathered photos, and collected artifacts. These scattered sources of information were greeted with acclaim when he returned to Paris because French anthropology lagged behind that in England and the French were eager to improve their competitive position. It was also relevant that French intellectuals favored Lévi-Strauss's abstract style of thought. Academic scholars in most other European nations held a more critical attitude toward this form of writing. Lévi-Strauss, who was pleasantly surprised by his unexpected celebrity, decided that anthropology might be the best route to fame and fortune. I suspect that if he had grown up in Munich, Moscow, Stockholm, or London, his brief encounters with the Brazilian Indians would not have attracted as much admiration and he would have selected a different field of inquiry that might not have brought him the international stature he enjoyed during the second half of his long life.

(Continues...)

Excerpted from Psychology's Ghostsby Jerome Kagan Copyright © 2012 by Yale University. Excerpted by permission of Yale UNIVERSITY PRESS. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.