40 Penelitian berpengaruh di psikologi

50 Chapter I The Biological Basis of Human Behavior worst performance on the visual cliff was the turtle. The baby turtles chosen to be tested were of the aquatic variety because the researchers expected that they might prefer the deep side of the “cliff” because their natural environ- ment is water. However, it appeared that the turtles were “smart” enough to know that they were not in water: 76% of them crawled off onto the shallow side, while 24% went “over the edge”: “The relatively large minority that chose the deep side suggests either that this turtle has poorer depth perception than other animals, or its natural habitat gives it less occasion to ‘fear’ a fall” (p. 67). Clearly, if you live your life in water, the survival value of depth perception, in terms of avoiding falls, would be diminished. Gibson and Walk pointed out that all of their observations were consistent with evolutionary theory. That is, all species of animals, if they are to survive, need to develop the ability to perceive depth by the time they achieve inde- pendent movement. For humans, this does not occur until around 6 months of age; for chickens and goats it is nearly immediate (by 1 day old); and for rats, cats, and dogs, it is about 4 weeks of age. The authors conclude, therefore, that this capacity is inborn because to learn it through trial and error would cause too many potentially fatal accidents. If we are so well prepared biologically, why do children take so many falls? Gibson and Walk explained that human infants’ perception of depth had matured sooner than had their skill in movement. During testing, many of the infants supported themselves on the deep side of the glass as they turned on the center board, and some even backed up onto the deep side as they began to crawl toward the mother across the shallow side. If the glass had not been there, some of the children would have fallen off the “cliff”! criTicisMs and subsequenT research The most common criticism of the researchers’ conclusions revolves around the question of whether they really proved that depth perception is innate in humans. As mentioned, by the time infants were tested on the visual cliff, they had already learned to avoid such situations. A later study placed younger infants, ages 2 to 5 months, on the glass over the deep side of the visual cliff. When this happened, all the babies showed a decrease in heart rate. Such a decrease is thought to be a sign of interest, not fear, which is accompanied by heart rate increases (Campos et al., 1978). This indicates that these younger infants had not yet learned to fear the drop-off and would learn the avoidance behavior somewhat later. These findings argued against Gibson and Walk’s position. It is important to notice, however, that although there was and still is controversy over just when we are able to perceive depth (the nativists vs. the empiricists), much of the research that is done to find the answer incorporates the visual cliff apparatus developed by Gibson and Walk. In addition, other related research using the visual cliff has turned up some fascinating findings. One example is the work of Sorce et al. (1985), who put 1-year-old infants on a visual cliff for which the drop-off was neither shallow nor deep

Reading 4 Watch out for the Visual Cliff! 51 but in between (about 30 inches). As a baby crawled toward the “cliff,” it would stop and look down. On the other side, as in the Gibson and Walk study, the mother was waiting. Sometimes the mother had been instructed to maintain an expression of fear on her face, while other times the mother looked happy and interested. When infants saw the expression of fear, they refused to crawl any farther. However, most of the infants who saw their mother looking happy checked the “cliff” again and crawled across. When the drop-off was made flat, the infants did not check with the mother before crawling across. This method of nonverbal communication used by infants in determining their behavior is called social referencing. recenT aPPlicaTions Gibson and Walk’s groundbreaking invention of the visual cliff still exerts a major influence on current studies of human development, perception, emotion, and even mental health. The following is a brief sample. A study by Berger and Adolph (2003) cited Gibson and Walk’s early study in their research on how toddlers analyze the characteristics of tasks involving heights, specifically crossing over a bridge. The researchers coaxed very young toddlers (16 months) to cross bridges of various widths, some with handrails, some without. They found that the children were significantly more likely to cross wider bridges than narrower ones (pretty smart for 16 months!). More interesting, however, was the finding that the toddlers were more likely to attempt the narrow bridge if it had handrails: “Infants who explored the bridge and handrail before stepping onto the bridge and devised alternative bridge-crossing strategies were more likely to cross successfully. [These] results challenge traditional conceptualizations of tools: babies used the handrail as a means for augmenting balance and for carrying out an otherwise impossible goal-directed task” (p. 594). How about using the concept of a technological visual cliff to help people with severe acrophobia (an irrational fear of heights)? This is a very common phobia affecting about 1 in 20 people. If you are a therapist and would like to have your acrophobic clients “face their fear of heights,” you will not have very many customers. Why? Because they are too scared! When someone has a true phobia he or she will do anything to avoid the feared object or situation. One study incorporating Gibson and Walk’s study found that the therapeutic effec- tiveness of a virtual drop, while still eliciting fear in the acrophobe, found it was less threatening for the client, which allowed for more effective therapy to reduce the irrational fear without ever leaving the therapist’s office (Coelho et al., 2009). conclusion Through the inventiveness of Gibson and Walk, behavioral scientists have been able to study depth perception in a clear and systematic way. Behavioral scientists continue to debate the question of whether this and other perceptual

52 Chapter I The Biological Basis of Human Behavior abilities are innate or learned. The truth may lie in a compromise that proposes an interaction between nature and nurture. Perhaps, as various studies have indicated, depth perception is present at birth, but fear of falling and avoidance of danger are learned through experience after the infant is old enough to crawl around enough to “get into trouble.” But whatever the questions are, elegant methodological advances such as the visual cliff allow us to continue to search for answers. Berger, S., & Adolph, K. (2003). Infants use handrails as tools in a locomotor task. Developmental Psychology, 39, 594–605. Campos, J., Hiatt, S., Ramsay, D., Henderson, C., & Svejda, M. (1978). The emergence of fear on the visual cliff. In M. Lewis & L. A. Rosenblum (Eds.), The development of affect. New York: Plenum Press. Coelho, C., Waters. A., Hine, T., & Wallis, G. (2009). The use of virtual reality in acrophobia research and treatment. Journal of Anxiety Disorders, 23, 563–574. Sorce, J., Emde, R., Campos, J., & Klinnert, M. (1985). Maternal emotion signaling: Its effect on the visual cliff behavior of 1-year-olds. Developmental Psychology, 21, 195–200.

Chapter II ConsCiousness and the senses Reading 5 Take a Long Look Reading 6 To SLeep, no DoubT To DReam Reading 7 aS a CaTegoRy, IT’S a naTuRaL Reading 8 aCTIng aS If you aRe HypnoTIzeD The study of perception and consciousness is of great interest to psychologists because these activities define and reveal much of your psychological inter- action with your environment. Think for a moment about how your senses are bombarded constantly by millions of pieces of information from the com- bined stimuli that surround you at any given moment. It is impossible for your brain to process all of it, so your brain organizes this barrage of sensory data into sets of information that yield form and meaning. That’s what psychologists refer to as perception. Clearly, your level of consciousness, also commonly referred to as your state of awareness, governs to a large extent what you perceive and how your brain organizes it. As you go through your day, night, week, year, and life, you experi- ence many and varied states of awareness: You concentrate (or not), daydream, fantasize, sleep, dream; maybe you’ve been hypnotized at some point or used psychoactive drugs (even caffeine and nicotine are psychoactive drugs!). These varying mental conditions are all altered states of consciousness that produce changes in your perceptions of the world that, in turn, influence your behavior. Within the research areas of perception and consciousness, some of the most influential and interesting studies have focused on perceptual abilities in early childhood, sleep, dreams, and hypnosis. This section begins with a famous and influential study that contributed a brilliant and remarkable method that allows researchers to study the thinking processes, the perceptions, of preverbal infants as young as a few days old. This method, called preference looking, provides insights into the functioning of infants’ brains and how they conceptualize the world. The second reading contains two articles that changed psychology because they (1) discovered rapid eye movement (REM) sleep and (2) revealed the relationship between REM and dreaming. Third is an influential and controversial study proposing that dreams are not mysteri- ous messages from your unconscious, as Freud and others suggested (and as you probably believe), but rather that dreams are the result of purely random, electrochemical impulses firing off in your brain while you sleep. Fourth is 53

54 Chapter II Consciousness and the Senses one of many studies that have influenced traditional psychological thinking by making a case against the widespread belief that hypnosis is a unique and pow- erful altered state of consciousness. This last study offers evidence suggesting that hypnotized people are no different from normally awake people—they are just a bit more motivated to behave in certain ways. Reading 5: take a Long Look Fantz, R. L. (1961). The origin of form perception. Scientific American, 204(May), 61–72. If you want to know about other people’s perceptions of the world around them, an easy way to find out is to ask them. Depending, of course, on exactly what you ask, they will often tell you. But have you ever tried to ask this of an infant? As much as infants may seem, at times, to be trying to tell you what they are thinking and perceiving, they cannot. They can’t talk; they probably could not tell you very much if they could; and, most likely, they couldn’t even understand your question! If you have had the opportunity to spend time around infants (and you likely have to varying degrees), you may have often thought to yourself, “I wonder what this baby is thinking!” or “If only this baby could talk . . . .” Unfortunately, that’s not going to happen ( John Travolta’s series of Look Who’s Talking movies aside). But psychologists’ interest in studying and understand- ing infants has been a top priority throughout psychology’s history (this book contains seven studies that have focused on infants). However, in Robert Fantz’s discoveries, which we will discuss in this chapter, the questions that plagued the researchers were “How can we study an infant’s cognitive processes?” and “How can we catch a real glimpse inside very young babies’ brains to see what might be going on, what they are perceiving, and how much they really understand?” In the 1950s, Robert L. Fantz, a psychologist at Western Reserve University in Cleveland (now, Case Western Reserve University), noticed something very interesting about infants; however, these were not human infants but newly hatched chicks—that’s right, chickens. Fantz reported that almost immediately upon breaking out of their shell, chicks perceive their environment well enough to begin searching and pecking for food. (See “Watch Out for the Visual Cliff!” in the previous group of readings for more about the perceptual talents of chicks.) This suggested to Fantz that chicks, in some ways, actually have superior perceptual abilities than human infants, making the chicks ideal subjects for research in this area. That said, it is important to note that when psychologists study nonhuman animals, their ultimate goal is to apply what they learn to our understanding of human behavior, but we will further discuss that issue later. TheoreTical ProPosiTions Prior to Fantz’s studies, research had clearly demonstrated that human infants are able to perceive the world around them in some rudimentary ways, such as the ability to see light, discriminate basic colors, and detect movement.

Reading 5 Take a Long Look 55 However, as Fantz pointed out, “It has often been argued that they cannot respond to such stimuli as shape, pattern, size, or solidity; in short, they cannot perceive form” (p. 66). But Fantz was skeptical of this argument, so in the late 1950s and early 1960s he set about developing a new research technique that would allow researchers to study in greater detail what infants can perceive, to pinpoint when perceptual skills develop, and to determine the degree of com- plexity of their perceptual skills. He proposed that human infants, from the moment of birth, not entirely unlike newly hatched chicks, are actually able to perceive various forms, and this can be demonstrated by observing how babies “analyze” their world—that is, what they look at and for how long they look at it. This method of studying infants’ mental abilities, called preferential looking, swept through the psychology world and began a revolution, which continues today, into understanding the minds of infants. MeThod It wasn’t difficult for Fantz to demonstrate some of what newly hatched chicks could and could not perceive. Fantz simply presented the chicks, before they had any experience pecking for real food, with objects of different shapes and sizes and recorded how often they pecked at each one. They pecked signifi- cantly more often at round shapes versus pyramid shapes; circles more than triangles; spheres more than flat disks; and when shapes of various sizes of 1 circles were presented, they preferred those that were about 8 inch in diameter over larger or smaller sizes. Without any previous learning, chicks were able to perceive form, and they clearly preferred shapes most like potential food: seeds or grain. Fantz expressed in his article what you are probably thinking right now: “Of course, what holds true for birds does not necessarily apply to human beings” (p. 67). He considered the possibility that this innate ability in birds to perceive form (and this is true of many bird species) may not have developed during the evolution of primates (including humans), or that perhaps primates acquire such abilities only after a period of development or learning following birth. So, when Fantz turned his attention to primate infants, he needed a new research method because, obviously, primate infants do not peck at anything, and they don’t have the motor development to do so even if they are so inclined (which they aren’t because infants are not terribly fond of grain and seeds). Infants do engage in one behavior, however, that might allow them to be tested in a similar way to the chicks: They stare at things. If Fantz could figure out a way to see if they stare at some forms predictably more often or longer than others, the only explanation would be that they could tell the difference—that they could perceive form. Working at first with infant chimpanzees, the primate genetically most closely related to humans, Fantz and his associates developed what he called a “looking chamber,” which was basically a padded, comfortable bassinette inside of a large, plain box. In the top panel of the box were two open- ings for presenting objects to the infants and peepholes allowing the researchers to observe the looking behavior of the infants. When the researchers ascertained that infant chimps appeared to show a systematic preference for certain objects

56 Chapter II Consciousness and the Senses 0 5 10 15 20 Figure 5-1 Infants’ interest Average Seconds of Fixation in 1-Minute Test in form pairs as a function of average looking time for 220 tests. (Source: Reproduced with Permission. Copyright © 1961 Scientific American, Inc. All rights reserved.) over others (determined by duration of staring), they applied the same basic techniques to studying human babies. The researchers did nothing to interfere with the babies’ usual schedule or activities but simply placed the infants into the comfortable, padded view- ing box and presented various pairs of object for them to look at. The infants ranged in age from 1 to 15 weeks of age. The stimuli presented to the babies included solid and textured disks, spheres, an oval with a human face, an oval with the features of a human face jumbled up, and shapes and patterns of varying complexity (see Figure 5-1). The researchers revealed the objects in various paired combinations and observed the total amount of time during each 1-minute trial the infants spent staring at the different pairs of objects, as well as which object within each pair they “preferred” (stared at longer). Their findings provided powerful evidence that babies of all ages possess the ability to perceive and discriminate among diverse forms. resulTs For their first round of testing, the babies saw pairs of various black-and-white test patterns, including a square with horizontal stripes and a square with a bull’s-eye; a checkerboard and a plain, patternless square; a wide plus sign and a circle; and a pair of identical triangles as control stimuli. The results are graphically illustrated in Figure 5-1. Clearly the infants “preferred” the forms

Reading 5 Take a Long Look 57 with the greatest complexity (the bull’s-eye, stripes, and checkerboard). This degree of preference was the same, regardless of the infant’s age, which indicates that the ability to discriminate among these forms is innate—present at birth. Beginning at approximately 8 weeks of age, the infants preferred the bull’s-eye to the stripes and the checkerboard to the plain square. This time delay implies that either some learning has occurred in those 2 months or that maturation of the brain and/or visual system accounted for the change. As interesting as these findings were, an important link between the infants’ abilities and the earlier studies of the chicks was still missing. If human infants are born with an unlearned, natural ability to discriminate form, we must ask why. For chicks, the answer appears rather straightforward: They perceive the forms that allow them to find nourishment and to survive. How could such an innate ability to perceive specific forms have survival value for human infants? Maybe it is for a similar reason. Fantz wrote the following: In the world of the infant, people have an importance that is perhaps compara- ble to the importance of grain in the chick’s world. Facial pattern is the most distinctive aspect of a person . . . for distinguishing a human being from other objects and identifying him. So, a facelike pattern might be expected to bring out selective perception in an infant if anything could. (p. 70) In other words, human infants do not depend upon form perception for nourishment and survival; they depend on other people to care for them. Just as chicks can perceive specific shapes best, it would make sense that infants’ perceptual tendencies should favor the human face. And they do. Fantz’s team presented 49 infants between 4 days and 6 months old with three identically sized oval disks. One was painted with the features of a human face; another with those same features scrambled; and the third, the control disk, an oval with just a patch of black at one end equal to the total area of the facial features on the other two disks (see Figure 5-2). The infants clearly showed greater interest in the ovals with the facial features and stared at them intently while virtually ignoring the control oval. Moreover, this pref- erence was approximately the same strength for all infants regardless of age, demonstrating again that basic form perception is present at birth and ruling out a learning or developmental factor. abc Figure 5-2 Fantz’s Facial Figure Test. Infants preferred A over B, and strongly preferred A and B over C. (Source: Reproduced with Permission. Copyright © 1961 Scientific American, Inc. All rights reserved.)

58 Chapter II Consciousness and the Senses Wants undisturbed sleep: unconscious, unacceptable wishes threaten to disrupt sleep and create anxiety Wants undisturbed sleep: unconscious, unacceptable wishes threaten to disrupt sleepand createundisturbed sleepanxiey Red White Yellow 10 20 30 40 50 0 Percent of Total Fixation Time Figure 5-3 Infants’ looking time for patterns and colors (black bars = 8–12 months; grey bars = over 12 months of age). (Source: Reproduced with Permission. Copyright © 1961 Scientific American, Inc. All rights reserved.) In the final study reported in this article, the researchers tested the human infants again for their ability to recognize facial forms. The infants were presented with six flat disks, each 6 inches in diameter with the following designs: (1) a human face; (2) a bull’s-eye; (3) a random fragment of a printed page (such as a newspaper or textbook); (4) an entirely red disk; (5) an entirely fluorescent yellow disk; and (6) a plain white disk. The time of the infants’ first look at each disk was recorded. Which one do you think they looked at the most? If you said “the face,” you are correct; they gazed at the human face disk far more than any other form or color (see Figure 5-3). subsequenT research and recenT aPPlicaTions This study, like so many in this book, significantly changed psychology for two reasons: the groundbreaking discoveries and the method the researcher developed to make those discoveries possible. Until the middle of the 20th century, many behavioral and biomedical researchers assumed that babies were born with few if any perceptual or sensory abilities and that they devel- oped or learned most, if not all, of these skills as they interacted with their environment over time. This idea of the psychologically “empty” newborn was relatively easy to accept because we did not, at the time, possess the necessary research methodologies to reveal very young infants’ true capabilities. Fantz gave us preferential-looking methods that, quite literally, opened the doors to the mind of the infant. These methods are used so commonly today that they are to psychology what a microscope is to biology: one of the first tools researchers turn to when they want to study how babies think. Of course, the discovery that infants come into the world with various perceptual skills does

Reading 5 Take a Long Look 59 not reduce the importance of learning and development. But the inborn skills researchers have discovered using Fantz’s methods appear to set the stage for an infant’s future survival and growth. As Fantz points out, Innate knowledge of the environment is demonstrated by the preference of newly hatched chicks for forms likely to be edible and by the interest of young infants in kinds of forms that will later aid in object recognition, social respon- siveness, and spatial orientation. This primitive knowledge provides a founda- tion for the vast accumulation of knowledge through experience. (p. 72) Fantz’s discoveries ignited a research revolution into the perceptual abilities of infants. You can see the influence of Fantz’s methodological ingenuity throughout the fields of developmental and cognitive psychology. For exam- ple, some of the leading researchers in the world in the area of infant cogni- tion, such as Renee Baillargeon at the University of Illinois’s Infant Cognition Lab and Elizabeth Spelke at Harvard’s Laboratory for Developmental Studies, have made extensive use of Fantz’s preferential-looking research strategies in many studies (see Talbot, 2006, for a review of this work). In addition, Fantz’s work helped clarify when and how well babies can perceive depth and drop-offs as studied in greater detail by Gibson and Walk in their classic research incorporating the visual cliff (see Chapter I). Probably the most important extension of Fantz’s work is credited to Frances Horowitz at the University of Kansas, who discovered that in addition to preferential looking, babies also become bored seeing the same stimulus over and over (Horowitz, & Paden et al., 1972). When you show infants a novel visual pattern (such as those used in Fantz’s studies), they gaze at it for a given amount of time, but as you repeatedly present the same stimulus, the amount of time they look predictably decreases. This is called habituation. If you then change or alter the pattern, their interest appears to revive and they look at it longer, a response known as dishabituation. By combining preferential looking, habitua- tion, and dishabituation methodologies, researchers can now learn a great deal about what very young infants, even newborns, “know” about their world. For example, in a recent study, researchers wanted to see when humans acquire the ability to distinguish between “possible” objects and “impossible” objects (Shuwairi, Albert, & Johnson, 2007). You undoubtedly have seen so-called impossible objects that we often refer to as optical illusions. Figure 5-4 Figure 5-4 Babies can distinguish between a possible (a) and impossible (b) object at 4 months old.

60 Chapter II Consciousness and the Senses exemplifies the difference between a possible and impossible object. You looked longer at the impossible one, didn’t you? So do babies. Using preferential- looking and duration-of-gaze methods, the researchers found that infants as young a 4 months old indicate an awareness of the difference in that they stared at the impossible object longer, as if to say, “I can see something’s wrong with this object and I need to try to figure it out!” This is just a sample of hundreds of studies conducted every year by devel- opmental psychologists and other behavioral scientists whose fundamental methodologies rest on Robert Fantz’s discoveries. These methods are allowing us to peek inside the minds of infants as never before to see what they perceive and how they think. Virtually every time we take another look, we discover that they are “smarter” and perceive more of their world than we ever expected. Horowitz, F. D., Paden, L., Bhana, K., & Self, P. (1972). An infant-controlled procedure for study- ing infant visual fixations. Developmental Psychology, 7, 90. Shuwairi, S., Albert, M., & Johnson, S. (2007). Discrimination of possible and impossible objects in infancy. Psychological Science, 18(4), 303–307. Talbot, M. (2006, September 4). The baby lab. The New Yorker, 82(27), 91–101. Reading 6: to sLeep, no doubt to dReam . . . Aserinsky, E., & Kleitman, N. (1953). Regularly occurring periods of eye mobility and concomitant phenomena during sleep. Science, 118, 273–274. Dement, W. (1960). The effect of dream deprivation. Science, 131, 1705–1707. As you can see, this section is somewhat different from the others in that two articles are discussed; this is because the first study discovered a basic phe- nomenon about sleeping and dreaming that made the second study possible. The primary focus is William Dement’s work on dream deprivation, but to prepare you for that, Aserinsky’s findings must be addressed first. In 1952, Eugene Aserinsky, although a graduate student, was studying sleep. Part of his research involved observing sleeping infants. He noticed that as these infants slept, active eye movements occurred periodically. During the remainder of the night, only occasional slow, rolling eye movements occurred. He theorized that these periods of active eye movements might be associated with dreaming. However, infants could not tell him whether they had been dreaming or not. To test this idea, he expanded his research to include adults. Aserinsky and his coauthor, Nathaniel Kleitman, employed 20 normal adults to serve as participants. Sensitive electronic measuring devices were con- nected by electrodes to the muscles around the eyes of these participants. The leads from these electrodes stretched into the next room, where the participants’ sleep could be monitored. The participants were then allowed to fall asleep normally (participants participated on more than one night each). During the night, participants were awakened and interrogated, either during periods of eye activity or during periods when little or no eye movement was observed. The idea was to wake the participants and ask them if they had been dreaming and if they could remember the content of the dream. The results were quite revealing.

Reading 6 To Sleep, No Doubt to Dream . . . 61 For all the participants combined, a total of 27 awakenings were done during periods of sleep accompanied by rapid eye movements. Of these, 20 reported detailed visual dreams. The other 7 reported “the feeling of having dreamed” but could not recall the content in detail. During periods of no eye movement, 23 awakenings were instigated; in 19 of these instances, the partici- pants did not report any dreaming, while in the other four, the participants felt vaguely as if they might have been dreaming, but they were not able to describe any dreams. On some occasions, participants were allowed to sleep through the night uninterrupted. It was found that the latter group experienced between three and four periods of eye activity during the average of 7 hours of sleep. Although it may not have seemed so remarkable at the time, Aserinsky had discovered what is very familiar to most of us now: rapid eye movement (REM) sleep, or dreaming sleep. From his discovery grew a huge body of research on sleep and dreaming that continues to expand. Over the years, as research methods and physiological recording devices have become more sophisticated, we have been able to refine Aserinsky’s findings and unlock many of the mysteries of sleep. For example, we now know that after you fall asleep, you sleep in four stages, beginning with the lightest sleep (Stage 1) and progressing into deeper and deeper stages. After you reach the deepest stage (Stage 4), you begin to move back up through the stages: Your sleep becomes lighter and lighter. As you approach Stage 1 again, you enter REM, which is a very different kind of sleep. You do most of your dreaming during REM sleep. However, contrary to popular belief, research has revealed that you do not move around very much during REM. Your body is immobilized by electrochemical messages from your brain that paralyze your muscles. This is most likely an evolutionary survival mechanism that prevents you from acting out your dreams and possibly injuring yourself or worse. Following a short period in REM, you proceed back into the four stages of sleep called non–rapid-eye-movement sleep (NON-REM, or NREM). During the night, you cycle between NREM and REM about five or six times (your first REM period comes about 90 minutes after falling asleep), with NREM becom- ing shorter and REM becoming longer (thereby causing you to dream more toward morning). (By the way, everyone dreams. Although a small percentage of individuals never remember dreams, sleep research has determined that we all have them.) All this knowledge springs from the discovery of REM by Aserinsky in the early 1950s. One of the leading researchers who followed Aserinsky in giving us this wealth of information on sleeping and dreaming is William Dement of Stanford University. Around the same time of Aserinsky’s find- ings, Dement was beginning his decades of groundbreaking research into sleeping and dreaming. TheoreTical ProPosiTions What struck Dement as most significant was the discovery that dreaming occurs every night in everyone. As Dement states in his article, “Since there appear to be no exceptions to the nightly occurrence of a substantial amount

62 Chapter II Consciousness and the Senses of dreaming in every sleeping person, it might be asked whether or not this amount of dreaming is in some way a necessary and vital part of our existence” (p. 1705). This led him to ask some obvious questions: “Would it be possible for human beings to continue to function normally if their dream life were completely or partially suppressed? Should dreaming be considered necessary in a psychological sense or a physiological sense or both?” (p. 1705). Dement decided to try to answer these questions by studying participants who had somehow been deprived of the chance to dream. At first he tried using depressant drugs to prevent dreaming, but the drugs themselves produced too great an effect on the participants’ sleep patterns to allow for valid results. Finally, he decided on a novel method of preventing dreaming by waking participants up every time they entered REM sleep during the night. MeThod drasTic Dement’s article reported on the first eight participants in an ongoing sleep and dreaming research project. The participants were all males ranging in age from 23 to 32. A participant would arrive at the sleep laboratory around his usual bedtime. Small electrodes were attached to the scalp and near the eyes to record brain-wave patterns and eye movements. As in the Aserinsky study, the wires to these electrodes ran into the next room so that the participant could sleep in a quiet, darkened room. The procedure for the study was as follows: For the first several nights, the participant was allowed to sleep normally for the entire night. This was done to establish a baseline for each participant’s usual amount of dreaming and overall sleep pattern. Once this information was obtained, the next step was to deprive the participant of REM or dream sleep. Over the next several nights (the number of consecutive deprivation nights ranged from three to seven for the various participants), the experimenter would awaken the participant every time the information from the electrodes indicated that he had begun to dream. The participant was required to sit up in bed and demonstrate that he was fully awake for several minutes before being allowed to go back to sleep. An important point mentioned by Dement was that the participants were asked not to sleep at any other times during the dream study. This was because if participants slept or napped, they might dream, and this could con- taminate the findings of the study. Following the nights of dream deprivation, participants entered the recovery phase of the experiment. During these nights, the participants were allowed to sleep undisturbed throughout the night. Their periods of dream- ing continued to be monitored electronically, and the amount of dreaming was recorded as usual. Next, each participant was given several nights off (something they were very glad about, no doubt!). Then six of them returned to the lab for another series of interrupted nights. These awakenings “exactly duplicated the dream- deprivation nights in number of nights and number of awakenings per night. The only difference was that the participant was awakened in the intervals

Reading 6 To Sleep, No Doubt to Dream . . . 63 between eye-movement (dream) periods. Whenever a dream period began, the participant was allowed to sleep on without interruption and was awak- ened only after the dream had ended spontaneously” (p. 1706). Participants again had the same number of recovery nights as they did following the dream- deprivation phase. These were called control recovery and were included to elim- inate the possibility that any effects of dream deprivation were not due simply to being awakened many times during the night, whether dreaming or not. resulTs Table 6-1 summarizes the main findings reported. During the baseline nights, when participants were allowed to sleep undisturbed, the average amount of sleep per night was 6 hours and 50 minutes. The average amount of time the participants spent dreaming was 80 minutes, or 19.5% (see Table 1, column 1). Dement discovered in these results from the first several nights that the amount of time spent dreaming was remarkably similar from participant to participant. In fact, the amount of variation among the dreamers was only plus or minus 7 minutes! The main point of this study was to examine the effects of being deprived of dreaming, or REM, sleep. The first finding to address this was the number of awakenings required to prevent REM sleep during the dream-deprivation nights. As you can see in Table 6-1 (column 3a), on the first night, the experi- menter had to awaken the participants between 7 and 22 times in order to block REM. However, as the study progressed, participants had to be awak- ened more and more often in order to prevent them from dreaming. On the last deprivation night, the number of forced awakenings ranged from 13 to 30 (column 3b). On average, there were twice as many attempts to dream at the end of the deprivation nights. Table 6-1 Summary of Dream-Deprivation results paRTICIpanT 1. 2. 3a. 3b. 4. 5. numbeR of peRCenT numbeR awakenIngS peRCenT peRCenT DReam of DReam DReam DReam TIme: DepRIvaTIon fIRST LaST TIme: TIme: nIgHT nIgHT baSeLIne nIgHTS ReCoveRy ConTRoL 1. 19.5 58 14 34.0 15.6 2. 18.8 77 24 34.2 22.7 3. 19.5 5 11 30 17.8 20.2 4. 18.6 57 23 26.3 18.8 5. 19.3 5 10 20 29.5 26.3 6. 20.8 4 13 20 29.0 — 7. 17.9 4 22 30 19.8 (28.1)* 16.8 8. 20.8 39 13 — average 19.5 4.38 11 22 —** 20.1 26.6 * Second recovery night. ** participant dropped out of study before recovery nights. (adapted from data on p. 1707.)

64 Chapter II Consciousness and the Senses The next and perhaps most revealing result was the increase in dream- ing time after the participants were prevented from dreaming for several nights. The numbers in Table 6-1 (column 4) reflect the first recovery night. The average total dream time on this night was 112 minutes, or 26.6% (com- pared with 80 minutes and 19.5% during baseline nights in column 1). Dement pointed out that two participants did not show a significant increase in REM (participants 3 and 7). If they are excluded from the calculations, the average total dream time is 127 minutes, or 29%. This is a 50% increase over the average for the baseline nights. Although only the first recovery night is reported in Table 6-1, it was noted that most of the participants continued to show elevated dream time (compared with baseline amounts) for five consecutive nights. Now you’re thinking, “Wait a minute!” Maybe this increase in dreaming has nothing to do with REM deprivation at all. Maybe it’s just because these par- ticipants were awakened so often. You’ll remember that Dement planned for your astute observation. Six of the participants returned after several days of rest and repeated the procedure exactly, except they were awakened between REM periods (the same number of times). This produced no significant increases in dreaming. The average time spent dreaming after the control awakenings was 88 minutes, or 20.1% of the total sleep time (column 5). When compared to 80 minutes, or 19.5%, in column 1, no significant difference was found. discussion Dement tentatively concluded from these findings that we need to dream. When we are not allowed to dream, there seems to be some kind of pressure to dream that increases over successive dream-deprivation nights. This was evi- dent in his findings from the increasing number of attempts to dream follow- ing deprivation (column 3a vs. column 3b) and in the significant increase in dream time (column 4 vs. column 1). He also notes that this increase continues over several nights so that it appears to make up in quantity the approximate amount of lost dreaming. Although Dement did not use the phrase at the time, this important finding has come to be known as the REM-rebound effect. Several interesting additional discoveries were made in this brief, yet remarkable article. If you return to Table 6-1 for a moment, you’ll see that two participants, as mentioned before, did not show a significant REM-rebound effect (participants 3 and 7). It is always important in research incorporating a relatively small number of participants to attempt to explain these excep- tions. Dement found that the small increase in participant 7 was not difficult to explain: “His failure to show a rise on the first recovery night was in all like- lihood due to the fact that he had imbibed several cocktails at a party before coming to the laboratory, so the expected increase in dream time was offset by the depressing effect of the alcohol” (p. 1706). Participant 3, however, was more difficult to reconcile. Although he showed the largest increase in the number of awakenings during depriva- tion (from 7 to 30), he did not have any REM rebound on any of his five

Reading 6 To Sleep, No Doubt to Dream . . . 65 recovery nights. Dement acknowledged that this participant was the one exception in his findings and theorized that perhaps he had an unusually sta- ble sleep pattern that was resistant to change. The eight participants were monitored for any behavioral changes that they might experience due to the loss of REM sleep. All the participants devel- oped minor symptoms of anxiety, irritability, or difficulty concentrating during the REM interruption period. Five of the participants reported a clear increase in appetite during the deprivation, three of whom gained 3 to 5 pounds. None of these behavioral symptoms appeared during the period of control awakenings. significance of The findings and subsequenT research More than 40 years after this preliminary research by Dement, we know a great deal about sleeping and dreaming. Some of this knowledge was dis- cussed briefly and previously in this chapter. We know that most of what Dement reported in his 1960 article has stood the test of time. We all dream, and if we are somehow prevented from dreaming one night, we dream more the next night. There does indeed appear to be something basic in our need to dream. In fact, the REM-rebound effect can be seen in many animals. One of Dement’s accidental findings, which he reported only as a minor anecdote, now has greater significance. One way that people may be deprived of REM sleep is through the use of alcohol or other drugs, such as ampheta- mines and barbiturates. Although these drugs increase your tendency to fall asleep, they suppress REM sleep and cause you to remain in the deeper stages of NREM for greater portions of the night. For this reason many people are unable to break the habit of taking sleeping pills or alcohol in order to sleep. As soon as they stop, the REM-rebound effect is so strong and disturbing that they become afraid to sleep and return to the drug to avoid dreaming. An even more extreme example of this problem occurs with alcoholics who may have been depriving themselves of REM sleep for years. When they stop drink- ing, the onset of REM rebound may be so powerful that it can occur while they are awake! This may be an explanation for the phenomenon known as delirium tremens (DTs), which usually involve terrible and frightening halluci- nations during withdrawal (Greenberg & Perlman, 1967). Dement spent decades following up on his early preliminary findings regarding the behavioral effects of dream deprivation. In his later work, he deprived participants of REM for much longer periods of time and found no evidence of harmful changes. He concluded that “[a] decade of research has failed to prove that substantial ill effects result even from prolonged selective REM deprivation” (Dement, 1974). Research with its origins in Dement’s early work reported here suggests that a greater synthesis of proteins takes place in the brain during REM sleep than during NREM sleep. Some believe that these chemical changes may rep- resent the process of integrating new information into the memory structures of the brain and may even be the organic basis for new developments in personality (Rossi, 1973).

66 Chapter II Consciousness and the Senses recenT aPPlicaTions Most experts in the field of sleep and dreaming credit Aserinsky with the discovery of REM sleep. Studies relating to sleeping, dreaming, or sleep disorders attribute that basic fact to him. Consequently, his early work with Kleitman is frequently cited in many recent scientific articles. Dement’s extension of Aserinsky’s work continues to be referred to frequently in a wide range of research articles relating to sleep patterns. One such recent study made the remarkable discovery that humans may dream during NREM sleep more than we thought (Suzuki et al., 2004). Using daytime napping, during which we tend to enter NREM sleep sooner than during normal nighttime sleep, the researchers found that when participants were asked to report on dreams during naps consisting only of NREM sleep, they were frequently able to do so. However, the researchers also found that “dream reports from NREM naps were less notable in quantity, vividness, and emotions than those from REM naps” (p. 1486). Another study relating to Dement and Aserinsky’s foundational research contends that humans develop during REM sleep a kind of protoconscious, a basic biological form of brain organization necessary for normal conscious- ness (Hobson, 2009). This basic human brain development is thought to begin before birth and continues throughout childhood. Hobson’s research proposes that early REM sleep provides us with a virtual model of our waking world that assists us in carrying out the tasks of our normal life while awake. The theory might help explain two phenomena: why infants spend more time in REM sleep than do adults, and why the human brain insists on obtaining a minimum amount of REM sleep every night. conclusion In 2000, Dement, who continues to oversee a very active sleep medicine research program at Stanford University, published, The Promise of Sleep: A Pioneer in Sleep Medicine Explores the Vital Connection Between Health, Happiness and a Good Night’s Sleep. In this book, written for the nonscientist, Dement draws upon his four decades of research on sleep and applies his vast accumu- lation of knowledge to helping all of us understand the vital importance of quality sleep and how to achieve it. In his book, Dement (2004) describes us as a “sleep-sick society” and sets forth his goals as a sleep researcher: For most of my career . . . I have worked unceasingly to change the way society deals with sleep. Why? Because the current way, or nonway, is so very bad . . . It greatly saddens me to think about the millions, possibly billions, of people, whose lives could be improved if they understood a few simple principles. Changing the way society and its institutions deal with sleep will do more good than almost anything else I can conceive, or certainly that was ever remotely in my grasp to accomplish. (pp. 4–5) To learn more about Dement’s ongoing work at Stanford University’s Center for Human Sleep Research, see http://med.stanford.edu/school/ psychiatry/humansleep.

Reading 7 As a Category, It’s a Natural 67 Dement, W. C. (1974). Some must watch while some must sleep. San Francisco, CA: Freeman. Dement. W. C. (2000). The promise of sleep: A pioneer in sleep medicine explores the vital connection between health, happiness and a good night’s sleep. New York: Dell. Greenberg, R., & Perlman, C. (1967). Delirium tremens and dreaming. American Journal of Psychiatry, 124, 133–142. Hobson, J. (2009). REM sleep and dreaming: Towards a theory of protoconsciousness. Neuroscience, 10, 803–813. Rossi, E. I. (1973). The dream protein hypothesis. American Journal of Psychiatry, 130, 1094–1097. Suzuki, H., Uchiyama, M., & Tagaya, H., et al. (2004). Dreaming during non-rapid eye movement sleep in the absence of prior rapid eye movement sleep. Sleep, 27(8), 1486–90. Reading 7: as a CategoRy, it’s a natuRaL Rosch, Eleanor H. (1973). Natural categories. Cognitive Psychology, 4, 328–350. In the 1934 Shirley Temple movie, Stand Up and Cheer, the great film actor and dancer who went by the name of “Stepin Fetchit,” sat on the porch steps of an old house examining one of his old, beat-up pieces of footwear, and lamented philosophically, “Why’s a shoe called a shoe?” His character often wondered why things were called what they were called, and in various ways psychologists have wondered the same thing. The behavioral scientists who focus on these sorts of questions study human cognition (thinking) and perception (humans’ interpretation of the world around them). One of the basic building blocks of these areas of research is the idea of concepts. Concepts are mental representations of your experience of the world that allow you to classify objects (furniture, vegetables, animals, professions, shoes, etc.) according to the characteristics they have in common. Concepts are extremely useful because they allow you to group objects into categories for efficient processing of information. For example, you know that a certain piece of furniture is a chair because it fits your “concept” of a chair. Therefore, it is not necessary for you to learn that a specific chair is called a chair each time you see an unfamiliar style so long as it fits into your category for chairs. Because it has come up in our conversation here, you are now thinking of a chair (right?). What features comprise your “chair concept”? You proba- bly think of a chair as having legs, a seat, and a back to lean against. Even though some chairs violate your rules (recliners and rocking chairs don’t really have legs), they still fit into your chair category well enough. However, if you were to encounter a bean bag “chair” without knowing what is was, you probably would not call it a chair. In fact you might not be sure what to call it. The question that has most interested cognitive psychologists is as follows: Where do you get your categories for objects? The traditional or “classical” view that was widely accepted prior to 1970 held that categories are a function of the language we speak. In other words, categories exist because we have words for them. For example, we have a category for animals that lay eggs, fly, have feathers, and chirp; the category is “bird.” This traditional view maintained that if we did not have a word for bird, the category or concept for bird would not exist. Therefore, concepts and categories should vary from culture to culture due to variations in language. And there is evidence of this. A frequently cited

68 Chapter II Consciousness and the Senses example is that the Inuit who live in far northern latitudes have 12 words in their native language for “snow,” whereas in English there are only one or two. Obviously, the Inuit need greater flexibility in communicating about snow due to the climate in which they live, and this is reflected in their language. South Pacific Islanders have no word at all in their language for snow; there- fore, scientists have assumed that such a concept would not exist for them. For many years, this theory of the origin of concepts was taken for granted by scientists throughout the social sciences in psychology, anthropology, lin- guistics, and sociology. During the early 1970s, Eleanor Rosch, at the University of California at Berkeley, published a series of studies that challenged the classical view and turned the field of cognitive psychology upside down. She is considered to have revolutionized the study of categorization. She proposed that categories do not necessarily arise from the language, but exist naturally on their own, in relation to humans’ biological abilities of perception. Her landmark study presented here involved two separate experiments and some rather technical procedures. For the sake of clarity and space limita- tions, a summary of the first experiment reported in the article previously cited will be detailed here. TheoreTical ProPosiTions Rosch theorized that if the prevailing theory were correct, all objects belonging to a certain category would have approximately equal status in that category— that is, they would fit into it equally well. She observed, however, that this is not the case. Instead, some “members” of a category are perceived by people to be better examples of the category than others (she called these prototypes). As an example of this, consider again the category of “bird.” Now, quickly, picture a bird in your mind. You probably pictured something like a robin, blue jay, wren, or sparrow (maybe a crow or an eagle). It is quite unlikely that you thought immediately of a goose, a chicken, an ostrich, or a penguin. According to Rosch this is because a robin fits your prototype (or “ideal example”) of a bird better than a chicken. In other words, a robin exhibits all or most of the fea- tures that describe your category of bird and you, therefore, judged it higher in “birdiness.” Conversely, an ostrich has few of these features—it doesn’t fly, doesn’t chirp, is too big, etc.—and, consequently, an ostrich does not fit most people’s prototype for a bird very well at all. What Rosch argued was that most categories do not have clear bounda- ries as to what fits and what does not, but rather, our mental category borders are “fuzzy.” We decide if an object fits into a category by comparing it to our category prototypes. She also believed that categories can exist and are psycho- logically real even when there are no words in a person’s language with which to name them. To test this theory, in the late 1960s, Rosch traveled to New Guinea where a society of people live called the “Dani” (see the discussion of the study by Ekman in Reading 22 for other research within this country’s cultures). The Dani, until recently, existed, in essence, as a Stone Age culture and

Reading 7 As a Category, It’s a Natural 69 communicated in a language that did not include certain concepts that now exist in all modern cultures. Rosch’s early studies, including the one discussed here, focused on categories relating to color. English speakers use 11 major color categories: red, yellow, green, blue, black, gray, white, purple, orange, pink, and brown. Research has determined that speakers of English are able to agree on certain “focal colors”: those that are the best examples (or color prototypes) of each color category. For example, English speakers know “fire engine red” is the “focal color” for the category of red (you could say it is the “most” redlike) and it is identified as red much more quickly and easily than other “reddish” or off-red colors. More on this in a moment. . . . The Dani, however, only possess two color categories: “mili,” which describes dark, cool colors and “mola,” used for light, warm colors. Rosch theorized that if the original, “classical” view of categorization was correct— that language determines concepts—the Dani should possess only two color concepts. However, Rosch contended that all humans possess from birth— that is, have “preprogrammed” into our brains through evolution—many more than two color categories. To test this, she decided to teach the Dani new words for either eight focal (prototype) colors or eight nonfocal colors. She hypothesized that many more than two focal color categories were “psychologically” real for the Dani culture, even though names for them had never existed in their language. If this were true, focal colors (i.e., color prototypes) should be able to be learned by the Dani faster and easier than nonfocal colors. MeThod Participants The participants for her study were young Dani males who were all pretested to be sure no one was color blind. They were also tested to confirm that their knowledge of color terms was restricted to “mili” (dark) and “mola” (light). Interestingly, the Dani do not measure age, so based on size and general phys- ical maturity the researchers judged all the participants to be at approximately 12 to 15 years of age. The participants volunteered to join in the study and were paid for their participation. After they had completed the color-learning part of the procedure, they were divided into several groups of 12 participants each. The two most impor- tant experimental groups will be discussed here. color stimuli Glossy color chips, similar to those you might obtain from the paint store when you are repainting your house, were used as the stimuli for the colors to be learned. These chips, however, were developed scientifically to represent specific colors of exacting wave lengths. The color categories used were pink, red, yellow, orange, brown, green, blue, and purple. For one group of participants, the colors

70 Chapter II Consciousness and the Senses were the focal, prototype hue for each color (such as fire-engine red). These were the colors that Rosch theorized to be universally represented by natural categories and, therefore, easily identifiable, regardless of culture or language. For the other group, the hues of the eight color chips fell in between focal colors so that an English-speaker might call them “red-brown” or “yellow- green.” These were called ambiguous or “nonfocal colors.” Procedure The first challenge Rosch and her associates faced was to assign names in the Dani language for the various colors. This was not as easy as it might sound because the names needed to be words that were all equally frequent, familiar, and meaningful to the participants, so as to avoid built-in bias. Rosch discov- ered that the Dani use many names for what they call “sibs.” A sib was described by Rosch as a family group similar to a clan. (The English language uses the same word, which is a shortened version of siblings.) These names met her requirements for stand-ins for colors and so were used to represent the differ- ent color categories to be learned by the participants. To avoid a preference or bias, a participant’s own sib name was not used as a color category. Each participant was told that the task involved learning a new language that the experimenter would teach to him. At the beginning of the first day the colors were presented to the participant and the name assigned to each color was spoken by the researcher and repeated by the participant. Then the colors were shuffled and presented again. Each time the participant named the color correctly he was praised or, if he was incorrect, he was told the correct name. Over a period of five to twelve days, the participants were tested on their learning of the color categories and their progress was recorded until all the participants were able to name all the colors without error. Upon completion of the learning period, an additional task was per- formed by all participants to determine if this new ability was truly understood as a general concept and would transfer to new situations or was limited only to the specific colors learned. To test for this, each participant was shown a group of many colors and asked to identify eight of them that had not been part of the training categories. The success rate of this “transfer” task (determining if the learning would trans- fer to a different setting) was calculated for the participants in both groups. resulTs If humans possess inborn, preprogrammed abilities to perceive certain categories of colors (as Rosch hypothesized), then the results of the Dani’s learning task should demonstrate faster learning for the focal color categories than for the nonfocal colors, because focal colors are better prototypes of the color concept. Figure 7-1 summarizes the learning progress of the two groups over the testing period. The average number of errors over the entire learning period was 8.54 for the prototype color group compared to 18.96 for the ambiguous

Reading 7 As a Category, It’s a Natural 71 20 Average number of errors 15 Focal color subjects 10 Nonfocal color subjects 5 0 1 2 3 4 5 6 7 8 9 10 11 12 Test day Figure 7-1 Average rate of color category learning for the focal color and nonfocal color participants. (Based on data from p. 338.) color group. This difference was highly statistically significant. If you consult Figure 7-1, you can see that the group presented with the colors that most members of Western cultures consider central to a particular color category, were able to learn the names of the colors in only five days compared to eleven days for the group trying to learn the nonfocal color categories. Rosch felt that it was important to demonstrate that the skill of recogniz- ing color categories had been acquired in such a way that it would transfer to new situations—that it was not only a result of the study, but had become a “usable” concept. On the task in which participants were asked to identify colors that were not part of the learning, correct responses would be expected only about 12% of the time (the percentage correct due to chance alone) if the concept did not transfer. The participants in Rosch’s study correctly iden- tified the colors not used with 90% accuracy. One additional informal finding reported by Rosch was that four of the participants in the nonfocal color group became very frustrated during the learning period and wanted to quit before learning all the color names. It took a great deal of persuasion to convince them to continue until they completed the task. This problem was not encountered with the focal color group who generally seemed to enjoy the experimental process. All this may seem like a long way to go for rather simple findings, but, as mentioned at the beginning of this chapter, the results from this and additional studies by Rosch and others had a profound effect on our knowl- edge of how the brain works. First, Rosch’s discussion of her findings will be summarized followed by a brief glimpse of the huge amount of subsequent, related research.

72 Chapter II Consciousness and the Senses discussion Rosch had found a way to test a theory that, on the surface, would seem nearly impossible to test. Can you think of a concept or category of objects that does not exist for speakers of English (or any other language) in the way color categories are missing from the Dani language? There may be some, but they are difficult to find and even more difficult to test. The notion to locate and study a culture that does not acknowledge color cate- gories was ingenious in itself. But the weight of her contributions lies more in what she discovered. The main finding was that people from a culture that did not possess concepts for colors could learn colors that comprised hypothesized proto- types faster than nonprototype colors. This finding indicated that certain concepts exist in the brains of all humans regardless of the language they speak or whether they have ever used the concepts. This was a major discovery. Because these concepts appear to be part of the biological structure of humans, Rosch called them “natural categories” (the title of her article). The reason this study had such an impact on psychological research was that suddenly the nearly universally accepted idea that language produces con- cepts had been changed to the radically opposing view that linguistic concepts stem from and form around these naturally occurring categories. Rosch concludes her article by suggesting further implications of her findings: In short, the evidence which has been presented regarding the structure and learning of color categories may have implications beyond the domain of color: (a) there may be other domains which are organized into natural categories and (b) even in nonperceptual categories, artificial prototypes (the best examples of nonperceptual categories) once developed, may affect learning and process- ing of categories in that domain in a manner similar to the effects of natural prototypes. (p. 349) What this means, is that most of what you perceive is analyzed and categorized by your brain according to how well or poorly it matches an appropriate prototype (natural or not), rather than how well it meets the criteria of a formal linguistic definition. subsequenT research Various studies that followed Rosch’s early research supported the existence of natural categories and the use of prototypes in concept formation. Rosch and her colleagues as well as others expanded on the early findings reported in this chapter to demonstrate its broader implications. For example, she further demonstrated that concepts do not have the clear, distinct boundaries that might exist if we used a strict linguistic defini- tion to categorize objects, but rather, as mentioned earlier in this chapter, concepts are indeed fuzzy and somewhat overlapping (see Rosch, 1975). If we return once again to our example of your concept of “bird,” would you say a

Reading 7 As a Category, It’s a Natural 73 kiwi is a bird? How about a bat? You may have formal knowledge that a kiwi is a bird (even though it doesn’t fly, or chirp, or sit in trees), but when you think of a bird, a kiwi rarely comes to mind (well, maybe now it will!). On the other hand, you know that a bat is not a bird and yet it flies, makes a sort of a chirping sound, and some of them live in trees. So, some people may, on some level, conceive of bats as birds. As another example, consider your category for “fruit.” What fruits are you thinking of? Apples, oranges, or bananas are usually the ones named first. What about a tomato? A tomato may be a fruit, but it is a poor example (compared to your prototype) of one because it is quite distant in resemblance to your prototypical fruit. Remember, psychologists are interested more in how you think than whether you are technically correct. (By the way, a kiwi is not only a bad example of a bird; it is also a bad example of a prototype of a fruit!) recenT aPPlicaTions Various research techniques to reveal how people conceptualize the world around them have been developed since Rosch first demonstrated the exist- ence of natural categories with the Dani in New Guinea (for a complete dis- cussion see Rosch, 1978). One method simply asks participants (from any culture) to use a number scale (such as from one to ten) to rate how well an object fits into a certain category (meaning how well it matches your proto- type for that category). For the category “dog” a German Sheppard might rate 10, but a French Bulldog might get a 3 (this has nothing to do with the quality of the breed, just how “doggy” people think they are). Another research tech- nique uses reaction time to measure how well something fits into a mental category (e.g., Dovidio, 1986; Rosch & Mervis, 1975; Unyk, 1990). One of the ways this is done is that you, as a participant, see or hear a statement, such as, “A turkey is a bird” and then press a button for true or false as fast as you can. Findings from this line of research demonstrate that the closer the category example matches your prototype of the category, the faster you will respond. “A turkey is a bird” will produce a significantly slower reac- tion time than, “A robin is a bird.” We all know that a French bulldog is a dog, but you would be unlikely to think of this particular breed when asked about dogs because it does not fit your prototype of a dog very well (unless you own one!). (Katsai Tetiana/ Shutterstock)

74 Chapter II Consciousness and the Senses A third method involves asking participants to produce examples of category members either by listing them or making line drawings. In a given amount of time, the participants will produce a far greater number of the more representative members of a category. For example, if you are asked to draw pieces of furniture, you will probably draw a chair, a couch, or a table, before you will draw a hutch or a bookcase. Or, if asked to list human emo- tions, happiness and anger might come to mind faster than, say, confusion or rage (e.g., Fehr & Russell, 1984). Just as is true of many of the studies summarized in this book, Rosch’s discoveries relating to natural categories and prototypes changed psychology’s view of your use of concepts, but over the 40 years since her findings, other research has appeared that either expands or questions her results. For example, some research has suggested that, although Rosch’s prototype theory appears to be valid, this does not mean researchers have abandoned our use of strict linguistic definitions. For example, one 2010 study, although acknowledging Rosch’s study discussed here, reported that other studies, including one with participants from another Papua subculture, do not appear to base their color perception primarily on prototypes within categories, but rather on linguistic cues (Tylen et al., 2010). The “truth” appears to be (if we can ever actually know the truth) that people will invoke linguistic definitions when that level of precision is nec- essary. Returning to the category of fruit provides an excellent example. If someone says to you, “Would you like a piece of fruit?” you do not think, “Fruit: The ripened seed-bearing structure of a plant.” Instead you immedi- ately access your prototype for “fruit,” which is most likely something such as an apple, an orange, or a banana. You would be quite surprised if you answered, “Yes!” and someone tossed you a pine cone (which is, technically, a fruit from a pine tree)! However, sometimes a formal, linguistic definition of fruit might be useful, such as, say, when you are on a nature walk and come upon an unusual plant with strange objects growing on it. Even though the objects on this plant bear little resemblance to your fruit proto- type, your formal, linguistic definition might allow you to say, “Look! This plant is bearing fruit.” Dovidio, J. (1984). Concept of emotion viewed from a prototype perspective. Journal of Experimental Psychology: General, 113, 464–486. doi: 10.1037/0096-3445.113.3.464 Fehr, B., & Russell, J. (1986). Racial stereotypes: The contents of their cognitive representations. Journal of Experimental Social Psychology, 22, 22–37. Rosch, E., & Mervis, C. (1975). Family resemblances: Studies in the internal structure of catego- ries. Cognitive Psychology, 7, 573–605. Rosch, E. H. (1975). Cognitive representations of semantic categories. Journal of Experimental Psychology: General, 104, 192–233. Rosch, E. H. (1978). Principles of categorization. In E. Rosch & B. Lloyd (Eds.), Cognition and categorization. NJ: Lawrence Erlbaum. Tylen, E., Weed, E., Wallentin, M., Roepstorff, A., & Frith, C. (2010). Language as a tool for inter- acting minds. Mind & Language, 25, 3–29. Unyk, A. (1990). An information-processing analysis of expectancy in music cognition. Psychomusicology, 9, 229–240.

Reading 8 Acting as If You Are Hypnotized 75 Reading 8: aCting as if you aRe hypnotized Spanos, N. P. (1982). Hypnotic behavior: A cognitive, social, psychological perspective. Research Communications in Psychology, Psychiatry, and Behavior, 7, 199–213. The alterations in consciousness with which we are all most familiar are related to sleep and dreaming. Another phenomenon relating to altered states of consciousness is hypnosis. Most people see hypnosis as a mysterious and pow- erful process of controlling the mind. The phrases and words that surround hypnosis, such as going under and trance, indicate that it is commonly consid- ered to be a separate and unique state of awareness, different from both wak- ing and sleep. And many psychologists support this view to varying degrees. Nicholas Spanos (1942–1994), however, led an opposing view that hypnosis is, in reality, nothing more than an increased degree of motivation to perform certain behaviors and can be explained fully without invoking notions of trances or altered states. The beginnings of hypnosis are usually traced back to the middle of the 18th century, a time when mental illness was first recognized by some as stem- ming from psychological rather than organic causes. One of the many influen- tial individuals who helped bring psychology out of the realm of witchcraft and devil possession was Franz Anton Mesmer (1733–1815). He believed that “hysterical disorders” were a result of imbalances in a “universal magnetic fluid” present in the human body. During strange gatherings in his laboratory, soft music would play; the lights would dim; and Mesmer, costumed like Dumbledore (from the Harry Potter series of books and movies), would take iron rods from bottles of various chemicals and touch parts of afflicted patients’ bodies. He believed that these elements and chemicals would transmit what he called the “animal magnetism” into the patients and provide relief from their symptoms. Interestingly, history has recorded that in many cases this treatment appears to be successful (probably due to placebo effects). It is from Mesmer that we acquired the word mesmerize, and many believe that his treatment included some of the techniques we now associate with hypnosis. Throughout the history of psychology, hypnosis (named after Hypnos, the Greek god of sleep) has played a prominent role, especially in the treatment of psychological disorders, and it was a major component in Freud’s psychoana- lytic techniques. Ernest Hilgard (1904–2001) was at the forefront of modern researchers who support the position that hypnosis is an altered psychological state (see Hilgard, 1978; Kihlstrom, 1998). His and others’ descriptions of hypnosis have included characteristics such as increased susceptibility to suggestion, involuntary performance of behaviors, improvements in recall, increased intensity of visual imagination, dissociation (the psychological sepa- ration from a person’s current environmental reality), and analgesia (lowered sensitivity to pain). Until the 1970s, the idea that hypnosis is capable of produc- ing thoughts, ideas, and behaviors that would otherwise be impossible, and that it is an altered state of consciousness, has been virtually undisputed.

76 Chapter II Consciousness and the Senses However, it is the job of scientists to look upon the status quo with a critical eye and, whenever they see fit, to attempt to debunk common beliefs. Social psychologist Nicholas Spanos suggested that the major assumptions underlying hypnosis, as set forth by Hilgard and others, should be questioned. In this article Spanos wrote, “The positing of special processes to account for hypnotic behavior is not only unnecessary, but also misleading . . . . Hypnotic behavior is basically similar to other social behavior and, like other social behavior, can be usefully described as strategic and goal-directed” (p. 200). In other words, Spanos contended that hypnotized participants are actually engaging in voluntary behavior designed to produce a desired consequence. He further maintained that although such behavior may result from increased motivation, it does not involve an altered state of consciousness. TheoreTical ProPosiTions Spanos theorized that all the behaviors commonly attributed to a hypnotic trance state are within the normal, voluntary abilities of humans. He main- tained that the only reason people define themselves as having been hypno- tized is that they have interpreted their own behavior under hypnosis in ways that are consistent with their expectations about being hypnotized. Spanos viewed the process of hypnosis as a ritual that in Western cultures carries a great deal of meaning. Participants expect to relinquish control over their own behavior, and as the process of hypnotic induction develops, they begin to believe that their voluntary acts are becoming automatic, involuntary events. An example of this that Spanos offered is that voluntary instructions are given early in the hypnotic procedure to the participant, such as “Relax the muscles in your legs,” but later these become involuntary suggestions, such as “Your legs feel limp and heavy.” In collaboration with various colleagues and associates, Spanos devoted nearly a decade of research prior to this 1982 article, demonstrating how many of the effects commonly attributed to hypnotic trances could be explained just as readily (or even more simply) in less mysterious ways. MeThod This article does not report on one specific experiment but rather summarizes a group of studies conducted by Spanos and his associates prior to 1982, which were designed to support his position countering Hilgard’s contention (and the popular belief) that hypnosis is a unique state of consciousness. Most of the findings reported were taken from 16 studies in which Spanos was directly involved and that offered interpretations of hypnotically produced behavior other than the common assumption of a unique altered state of being. resulTs and discussion Spanos claimed that two key aspects of hypnosis lead people to perceive it as an altered state of consciousness. One is that participants interpret their behavior during hypnosis as caused by something other than the self, thus

Reading 8 Acting as If You Are Hypnotized 77 making their actions seem involuntary. The second aspect is the belief discussed previously that the “hypnosis ritual” creates expectations in participants, which in turn motivate them to behave in ways that are consistent with their expecta- tions. The findings of the research Spanos reports in this article focus on how these frequently cited claims about hypnosis may be drawn into question. The belief That behavior is involuntary As participants are being hypnotized, they are usually asked to take various tests to determine if a hypnotic state has been induced. Spanos claimed that these tests are often carried out in such a way as to invite the participants to convince themselves that something out of the ordinary is happening. Hypnotic tests involve suggestions, such as “Your arm is heavy and you cannot hold it up”; “Your hands are being drawn together by some force and you cannot keep them apart”; “Your arm is as rigid as a steel bar and you cannot bend it”; or “Your body is so heavy that you cannot stand up.” Spanos interpreted these test suggestions as containing two interrelated requests. One request asks partici- pants to do something, and the other asks them to interpret the action as hav- ing occurred involuntarily. Some participants fail completely to respond to the suggestion. Spanos claimed that these participants do not understand that they must voluntarily do something to initiate the suggested behavior and instead simply wait for their arms or body to begin to move. Other participants respond to the suggestion but are aware that they are behaving voluntarily. Still other participants agree to both requests; they respond to the suggestion and interpret their response as beyond their control. Spanos suggested that whether participants interpret their behavior to be voluntary or involuntary depends on the way the suggestion is worded. In one of his studies, Spanos put two groups of participants through a hypnosis induc- tion procedure. Then to one group he made various behavior suggestions, such as “Your arm is very light and is rising.” To the other group he gave direct instructions for the same behaviors, such as “Raise your arm.” Afterward he asked the participants if they thought their behaviors were voluntary or invol- untary. The participants in the suggestion group were more likely to interpret their behaviors as involuntary than were those in the direct instruction group. Right now, while you are reading this page, hold your left arm straight out and keep it there for a couple of minutes. You will notice that it begins to feel heavy. This heaviness is not due to hypnosis; it’s due to gravity! If you are hypnotized and given the suggestion that your outstretched arm is becoming heavy, it would be very easy for you to attribute your action of lowering your arm to involuntary forces (you want to lower it anyway!). But what if you are given the suggestion that your arm is light and rising? If you raise your arm, it should be more difficult to interpret that action as involuntary, because you would have to ignore the contradictory feedback provided by gravity. Spanos tested this idea and found that such an interpretation was more difficult. Participants who believed they were hypnotized were significantly more likely to define as involuntary their behavior of arm lowering than that of arm raising. In the traditional view of hypnosis, the direction of the arm in

78 Chapter II Consciousness and the Senses the hypnotic suggestion should not make any difference; it should always be considered involuntary. Suggestions made to hypnotic participants often ask them to imagine cer- tain situations in order to produce a desired behavior. If you were a participant, you might be given the suggestion that your arm is rigid and you cannot bend it. To reinforce this suggestion, it might be added that your arm is in a plaster cast. Spanos believed that some people may become absorbed in these imaginal strat- egies more than others, which could have the effect of leading them to believe that their response (the inability to move their arm) was involuntary. His reason- ing was that if you are highly absorbed, you will not be able to focus on infor- mation that alerts you to the fact that the fantasy is not real. The more vividly you imagine the cast, its texture and hardness, how it got there, and so on, the less likely you are to remember that this is only your imagination at work. If this deep absorption happens, you might be more inclined to believe that your rigid- arm behavior was involuntary when actually it was not. In support of this, Spanos found that when participants were asked to rate how absorbed they were in a suggested imagined scenario, the higher the absorption rating, the more likely they were to interpret their related behavior as occurring involuntarily. Spanos also noted that a person’s susceptibility to hypnosis correlates with his or her general tendency to become absorbed in other activities, such as books, music, or daydreaming. Consequently, these individuals are more likely to willingly cooperate with the kind of suggestions involved in hypnosis. creation of expectations in hypnotic Participants Spanos claimed that the beliefs most people have about hypnosis are adequate in themselves to produce what is typically seen as hypnotic behavior. He fur- ther contended that these beliefs are strengthened by the methods used to induce and study hypnosis. He cited three examples of research that demon- strated how people might engage in certain behaviors under hypnosis because they think they should, rather than because of an altered state of awareness. First, Spanos referred to a study in which a lecture about hypnosis was given to two groups of students. The lectures were identical except that one group was told that arm rigidity was a spontaneous event during hypnosis. Later both groups were hypnotized. In the group that had heard the lecture including the information about arm rigidity, some of the participants exhib- ited this behavior spontaneously, without any instructions to do so. However, among the participants in the other group, not one arm became rigid. According to Spanos, this demonstrated how people will enact their experi- ence of hypnosis according to how they believe they are supposed to behave. The second hypnotic event that Spanos used to illustrate his position involved research findings that hypnotized participants claim the visual imagery they experienced under hypnosis was more intense, vivid, and real than similar imaginings when not hypnotized. Here, in essence, is how these studies typically have been done: Participants are asked to imagine scenes or situations in which they are performing certain behaviors. Then, these same

Reading 8 Acting as If You Are Hypnotized 79 participants are hypnotized and again asked to visualize the same or similar situations (the hypnotized and nonhypnotized trials can be in any order). These participants generally report that the imagery in the hypnotized condi- tion was significantly more intense. Spanos and his associates found, however, that when two different groups of participants are used, one hypnotized and one not, their average intensity ratings of the visual imagery are approximately equal. The difference in the two methods is probably explained by the fact that when two different groups are tested, the participants do not have any- thing to use for comparison. However, when the same participants are used in both conditions, they can compare the two experiences and rate one against the other. Because participants nearly always rate the hypnotic imagery as more intense, this supports the idea that hypnosis is really an altered state, right? If you could ask Spanos, he would say, “Wrong!” In his view, the parti- cipants who participate in both conditions expect the ritual of hypnosis to produce more intense imagery, and, therefore, they rate it accordingly. The third and perhaps most interesting demonstration of hypnosis addressed by Spanos was the claim that hypnosis can cause people to become insensitive to pain (the analgesia effect). One way that pain can be tested in the laboratory without causing damage to the participant is by using the “cold pressor test.” If you are a participant in such a study, you would be asked to immerse your arm in ice water (0 degrees centigrade) and leave it there as long as you could. After the first 10 seconds or so, this becomes increasingly painful, and most people will remove their arm within a minute or two. Hilgard (1978) reported that participants who received both waking and hypnotic training in analgesia (pain reduction) reported significantly less cold-pressor pain during the hypnotized trials. His explanation for this was that during hypnosis, a person is able to dissociate the pain from awareness. In this way, Hilgard contended, a part of the person’s consciousness experiences the pain, but this part is hidden from awareness by what he called an “amnesic barrier.” Again, Spanos rejected a hypnotic explanation for these analgesic find- ings and offered evidence to demonstrate that reduction in perceived pain during hypnosis is a result of the participants’ motivation and expectations. All the research on hypnosis uses participants who have scored high on meas- ures of hypnotic susceptibility. According to Spanos, these individuals “have a strong investment in presenting themselves in the experimental setting as good hypnotic subjects” (p. 208). The participants know that a waking state is being compared to a hypnotic state and want to demonstrate the effectiveness of hypnosis. Spanos, working with his associate H. J. Stam, performed a similar study involving cold-pressor pain but with one major difference: Some partici- pants were told that they would first use waking analgesia techniques (such as self-distraction) and would then be tested using hypnotic pain-reduction methods, but other participants were not told of the later hypnotic test (see also Stam and Spanos, 1980). Figure 8-1 summarizes what Stam and Spanos found. When participants expected the hypnosis condition to follow the waking trials, they rated

80 Chapter II Consciousness and the Senses 100 80 60 Rating of pain intensity 40 20 0 Hypnotic Waking Hypnotic Waking Expectation of No expectation hypnosis of hypnosis Figure 8-1 Waking versus hypnotic analgesia: expectation versus no expectation. the analgesic effect lower in order to, as the authors state, “leave room” for improvement under hypnosis. Stam and Spanos claimed that this demon- strated how even the hypnotic behavior of pain insensitivity could be attributed to the participants’ need to respond to the demands of the situation rather than automatically assuming a dissociated state of consciousness. The most important question concerning all these findings reported by Spanos is whether we should reevaluate the phenomenon called hypnosis. And what does it mean if we were to decide that hypnosis is not the powerful mind-altering force that popular culture, and many psychologists, have portrayed it to be? iMPlicaTions of The findings In evaluating Spanos’s research, you should remember that his goal was not to prove that hypnosis does not exist but, rather, to demonstrate that what we call hypnotic behaviors are the result of highly motivated, goal-directed social behavior, not an altered and unique state of consciousness. It is well accepted

Reading 8 Acting as If You Are Hypnotized 81 among most behavioral scientists that people cannot be hypnotized against their will. Furthermore, under hypnosis, participants will not engage in acts they believe are antisocial, and they are not able to perform feats of super- human strength or endurance. In this article, Spanos has demonstrated how many of the more subtle aspects of hypnosis may be explained in less mysteri- ous and more straightforward ways than that of the hypnotic trance. What would be the implications of accepting Spanos’s contention that hypnosis does not exist? The answer to this question is “Perhaps none.” Whether the effects of hypnosis are produced by an altered state of awareness or by increased motivation does not change the fact that hypnosis is often a useful method of helping people improve something in their lives. One reason that there continues to be such widespread and unquestioning acceptance of the power of the hypnotic trance may be that humans need to feel that there is a way out, a last resort to solve their problems if all else fails—something so omni- potent that they can even change against their own resistance to such change. Whether or not hypnosis is an altered state of consciousness remains a highly controversial issue. But whatever hypnosis is, it is not the panacea most people would like to find. Several studies have shown that hypnosis is no more effective than other methods of treatment to help people stop abusing alcohol and tobacco, improve their memory, or lose weight (see Lazar & Dempster, 1981, for a review of this research). recenT aPPlicaTions If, given the chance, would you readily agree to be hypnotized or would you rather avoid it? A study based on Spanos’s research examined reluctant candi- dates for hypnosis (Capafons et al., 2005). The findings, as you will see, could be used either to dispute or confirm Spanos’s research. These hesitant parti- cipants, once convinced to agree to the hypnotic experience, were assigned to three groups: One group was given almost no information about what to expect from a hypnotic experience (the control group). Another group was given “cognitive-behavioral” information about the hypnotic experience— that is, what to expect about how they would think and behave while under hypnosis. The third group was told what a hypnotic trance was like: An altered state of consciousness (a trance) accompanied by cognitive dissociation (a sepa- ration of the consciousness from normal self-awareness). One might think that among those who were avoidant about being hypnotized to begin with, the cognitive-behavioral and the trance groups would be more resistant to the hypnotic procedure and the results of the hypnosis would be less obvious (or nonexistent). But, surprisingly, just the opposite was found. Both the cognitive-behavioral group and the trance group demon- strated greater hypnotic suggestibility than the control group. What do these findings mean? This is not particularly clear. They may indicate the old saying that “knowledge is power.” That is, people who receive more information about an “unknown” experience, feel a greater level of perceived control, and feel more able to “handle” themselves during the hypnotic procedure.

82 Chapter II Consciousness and the Senses Another study cited Spanos’s perspectives on hypnosis to question cer- tain therapeutic practices often employed by some psychotherapists to induce clients to recover ostensibly “repressed” memories of past sexual abuse (Lynn et al., 2003). The authors contended that hypnosis, along with other thera- peutic techniques, may distort memories or even create memories of abuse that never actually took place, especially in early childhood (see the reading on the work of Elizabeth Loftus in Chapter IV for more about recovered mem- ories). The researchers point out, based on Spanos’s research, that “adults’ memory reports from 24 months of age or earlier are likely to represent confabulations, condensations, and constructions of early events, as well as current concerns and stories heard about early events” (p. 42). In other words, the belief that hypnosis somehow allows clients to retrieve accurate memories of early traumatic experiences is misguided and may be subject to all the memory errors that exist in a nonhypnotized state. This, the authors contend, may in some cases, lead to false memories and accusations of abuse that never happened. Spanos elaborated his perspective on this potential misuse of hypnotic techniques in his 1994 book, Multiple Identities & False Memories: A Sociocognitive Perspective. conclusion Clearly, the debate goes on. Spanos continued his research until his untimely death in a plane crash in June 1994 (see McConkey & Sheehan, 1995). A sum- mary of his early work on hypnosis can be found in his 1988 book, Hypnosis: The Cognitive-Behavioral Perspective. Nicholas Spanos was a prolific and well- respected behavioral scientist who has been missed greatly by his colleagues and by all those who learned and benefited from his work (see Baker, 1994, for a eulogy to Nick Spanos). And, clearly, his research legacy will be carried on by others. His work on hypnosis changed psychology in that he offered an experimentally based, alternative explanation for an aspect of human con- sciousness and behavior that was virtually unchallenged for nearly 200 years. Baker, R. (1994). In memoriam: Nick Spanos. Skeptical Inquirer, 18(5), 459. Capafons, A., Cabañas, F., Alarcón, A., Espejo, B., Mendoza, E., Chaves, J., & Monje, A. (2005). Effects of different types of preparatory information on attitudes toward hypnosis. Contemporary Hypnosis, 22, 67–76. Hilgard, E. (1978). Hypnosis and consciousness. Human Nature, 1, 42–51. Kihlstrom, J. F. (1998). Attributions, awareness, and dissociation: In memoriam Kenneth S. Bowers, 1937–1996. American Journal of Clinical Hypnosis, 40(3), 194–205. Lazar, B., and Dempster, C. (1981). Failures in hypnosis and hypnotherapy: A review. American Journal of Clinical Hypnosis, 24(1), 48–54. Lynn, S., Loftus, E., Lilienfeld, S., & Lock, T. (2003). Memory recovery techniques in psycho- therapy: Problems and pitfalls. Skeptical Inquirer, 27, 40–46. McConkey, K., & Sheehan, P. (1995). Nicholas Spanos: Reflections with gratitude. Contemporary Hypnosis, 12, 36–38. Spanos, N. (1994). Multiple identities & false memories: A sociocognitive perspective. Washington, DC: American Psychological Association. Spanos, N., & Chaves, J. (1988). Hypnosis: The cognitive-behavioral perspective. New York: Prometheus. Stam, H. J., & Spanos, N. (1980). Experimental designs, expectancy effects, and hypnotic analgesia. Journal of Abnormal Psychology, 89, 751–762.

Chapter III CoNDItIoNINg AND leARNINg Reading 9 It’s Not Just About sAlIvAtINg Dogs! Reading 10 lIttle emotIoNAl AlbeRt Reading 11 KNocK WooD! Reading 12 see AggRessIoN . . . Do AggRessIoN! The area of psychology concerned with learning has produced a rather well- defined body of literature explaining the process underlying how animals and humans learn. Some of the most famous names in the history of psychol- ogy have made their most influential discoveries in this field—names that are easily recognized by those both inside and outside the behavioral sciences, such as Pavlov, Watson, Skinner, and Bandura. Picking a few of the most sig- nificant studies from this branch of psychology and from these researchers is no easy task, but the articles selected here can be found in nearly every introductory psychology textbook and are representative of the enormous contributions of these scientists. For Ivan Pavlov, we take a journey back to the early 1900s to review his work with dogs, metronomes, bells, salivation, and the discovery of the conditioned reflex. Second, John Watson, known for many contributions, is probably most famous (notorious?) for his 1920 ethically challenged experiment with Little Albert, which demonstrated for the first time how emotions could be shown to be a product of the environment rather than purely internal processes. For the third study in this section, we discuss B. F. Skinner’s famous demonstration of superstitious behavior in a pigeon and his explanation for how humans become superstitious in exactly the same way. Fourth, we examine the well-known “Bobo Doll Study,” in which Albert Bandura established that aggressive behaviors could be learned by children through their modeling of adult violence. Reading 9: It’s Not Just About sAlIvAtINg Dogs! Pavlov, I. P. (1927). Conditioned reflexes. London: Oxford University Press. Have you ever walked into a dentist’s office where the odor of the disinfectant made your teeth hurt? If you have, it was probably because the odor triggered an association that had been conditioned in your brain between that smell and your past experiences at the dentist. When you hear “The Star Spangled Banner” 83

84 Chapter III Conditioning and Learning played at the Olympic Games, does your heart beat a little faster? That happens to most Americans. Does the same thing happen when you hear the Italian national anthem? Unless you were raised in Italy, most likely it does not, because you have been conditioned to respond to one anthem but not to the other. And why do some people squint and become nervous if you inflate a balloon near them? It is because they have learned to associate the expanding balloon with something fearful (such as a loud pop! ). These are just a few of countless human behaviors that exist because of a process known as classical conditioning. The classical conditioning theory of learning was developed and articu- lated nearly a hundred years ago in Russia by one of the most familiar names in the history of psychology, Ivan Petrovich Pavlov (1849–1946). Unlike most of the research presented in this book, Pavlov’s name and his basic ideas of learning by association are widely recognized in popular culture (even a Rolling Stones’s song from the 1970s contained the line “I salivate like a Pavlov dog”). However, how Pavlov came to make his landmark discoveries and the true significance of his work are not so widely understood. Although Pavlov’s contributions to psychology were among some of the most important ever made, technically he was not a psychologist at all but, rather, a prominent Russian physiologist studying digestive processes. In 1904, his research on digestion earned him the Nobel Prize for science. Yet the dis- coveries that dramatically changed his career, and the history of psychology, began virtually by accident. In the late 1800s, psychology was a very young field of scientific study and was considered by many to be something less than a true science. Therefore, Pavlov’s decision to make such a radical turn from the more solid and respected science of physiology to the fledgling study of psychology was a risky career move. He wrote about this dilemma facing any scientist thinking about studying psychology in the early 1900s: It is logical that in its analysis of the various activities of living matter, physiology should base itself on the more advanced and more exact sciences, physics and chemistry. But if we attempt an approach from this science of psychology .  .  . we shall be building our superstructure on a science that has no claim to exactness . . . . In fact, it is still open to discussion whether psychology is a natural science, or whether it can be regarded as a science at all. (p. 3) Looking back on Pavlov’s discoveries, it was fortunate for the advancement of psychological science and for our understanding of human behavior that he took the risk and made the change. Pavlov’s physiological research involved the use of dogs as subjects for studying the role of salivation on digestion. He or his assistants would intro- duce various types of food or nonfood substances into a dog’s mouth and observe the rate and amount of salivation. To measure salivation scientifically, minor surgery was performed on the dogs so that a salivary duct was redi- rected through an incision in the dog’s cheek and connected to a tube that would collect the saliva. Throughout this research, Pavlov made many new and fascinating discoveries. For example, he found that when a dog received moist food, only a small amount of saliva would be produced, compared with a heavy

Reading 9 It’s Not Just About Salivating Dogs! 85 flow when dry food was presented. The production of saliva under these vary- ing conditions was regarded by Pavlov as a reflex, that is, a response that occurs automatically to a specific stimulus without the need for any learning. If you think about it, salivation is purely reflexive for humans, too. Suppose I ask you, as you read this sentence, to salivate as heavily as you can. You cannot do it. But if you are hungry and find yourself sitting in front of your favorite food, you will salivate whether you want to or not. As Pavlov continued his research, he began to notice strange events that were totally unexpected. The dogs began to salivate before any food reached their mouths and even before the odor of food was present. After a while, the dogs were salivating at times when no salivary stimulus was present at all. Somehow, the reflexive action of the salivary glands had been altered through the animals’ experience in the lab: “Even the vessel from which the food has been given is sufficient to evoke an alimentary reflex [of salivation] complete in all its details; and, further, the secretion may be provoked even by the sight of the person who has brought the vessel, or by the sound of his footsteps” (p. 13). This was the crossroads for Pavlov. He had observed digestive responses occurring to stimuli seemingly unrelated to digestion, and pure physiology could not provide an explanation for this. The answer had to be found in psychology. TheoreTical ProPosiTions Pavlov theorized that the dogs had learned from experience in the lab to expect food following the appearance of certain signals. Although these signal stimuli do not naturally produce salivation, the dogs came to associate them with food, and thus responded to them with salivation. Consequently, Pavlov determined two kinds of reflexes must exist. Unconditioned reflexes are inborn and automatic, require no learning, and are generally the same for all members of a species. Salivating when food enters the mouth, jumping at the sound of a loud noise, and the dilation of your pupils in low light are examples of unconditioned reflexes. Conditioned reflexes, on the other hand, are acquired through experience or learning and may vary a great deal among individual members of a species. A dog salivating at the sound of footsteps, or you feeling pain in your teeth when you smell dental disinfectant, are conditioned reflexes. Unconditioned reflexes are formed by an unconditioned stimulus (UCS) produc- ing an unconditioned response (UCR). In Pavlov’s studies, the UCS was food and the UCR was salivation. Conditioned reflexes consist of a conditioned stimulus (CS), such as the footsteps, producing a conditioned response (CR), salivation. You will notice that the response in both these examples is salivation, but when the saliva- tion results from hearing footsteps, it is the learning, and not the dog’s natural tendencies, that produced it. Pavlov wanted to answer this question: Conditioned reflexes are not inborn, so exactly how are they acquired? He proposed that if a particular stim- ulus in the dog’s environment was often present when the dog was fed, this stimulus would become associated in the dog’s brain with food; it would signal

86 Chapter III Conditioning and Learning the approaching food. Prior to being paired with the food, the environmental stimulus did not produce any important response. In other words, to the dogs, it was a neutral stimulus (NS). When the dogs first arrived at the lab, the assist- ant’s footsteps might have produced a response of curiosity (Pavlov called it the “What is it?” response), but hearing the footsteps certainly would not have caused the dogs to salivate. The footsteps, then, were a neutral stimulus. However, over time, as the dogs heard the same footsteps just prior to being fed every day, they would begin to associate the sound with food. Eventually, according to the theory, the footsteps alone would cause the dogs to salivate. Pavlov proposed that the process by which a neutral stimulus becomes a condi- tioned stimulus could be diagrammed as follows: Step 1 NS + UCS UCR (footsteps) + (food) (salivation) Step 2 UCS (Repeat step 2 several times) (food) UCR Step 3 (salivation) Step 4 CS (footsteps) CR (salivation) Now that he had a theory to explain his observations, Pavlov began a series of experiments to prove that it was correct. It is commonly believed that Pavlov conditioned dogs to salivate at the sound of a bell, which was true of his later studies. But as you will see, his early experiments involved a metronome. MeThod and resulTs Pavlov was able to build a special laboratory at the Institute of Experimental Medicine in Petrograd (which became Leningrad following Lenin’s death and has now returned to its original name of St. Petersburg) with funds donated by a philanthropic businessman from Moscow. This soundproof lab allowed for complete isolation of the subjects from the experimenters and from all extraneous stimuli during the experimental procedures. Therefore, a specific stimulus could be administered and responses could be recorded without any direct contact between the experimenters and the animals. After Pavlov had established this controlled research environment, the procedure was quite simple. Pavlov chose food as the unconditioned stimulus. As explained previously, food will elicit the unconditioned response of salivation. Then Pavlov needed to find a neutral stimulus that was, for the dogs, completely unrelated to food. For this he used the sound of the metronome. Over several conditioning trials, the dog was exposed to the ticking of the metronome and then was immediately presented with food: “A stimulus which was neutral of itself had been superimposed upon the action of the inborn alimentary reflex. We observed that, after several repetitions of the combined stimulation, the

Reading 9 It’s Not Just About Salivating Dogs! 87 sounds of the metronome had acquired the property of stimulating salivary secretion” (p. 26). In other words, the metronome had become a conditioned stimulus for the conditioned response of salivation. Pavlov and his associates elaborated on this preliminary finding by using different unconditioned and neutral stimuli. For example, they presented the odor of vanilla (NS) to the subjects prior to placing a lemon juicelike solution in the dog’s mouth (the UCS). The juice caused heavy salivation (UCR). After 20 repetitions of the pairing, the vanilla alone produced salivation. For a visual test, the dogs were exposed to an object that began to rotate just prior to the presentation of food. After only five pairings, the rotating object by itself (CS) caused the dogs to salivate (CR). The importance and application of Pavlov’s work extends far beyond salivating dogs. His theories of classical conditioning explained a major por- tion of human behavior and helped to launch psychology as a true science. significance of The findings The theory of classical conditioning (also called Pavlovian conditioning) is universally accepted and has remained virtually unchanged since its concep- tion through Pavlov’s work. It is used to explain and interpret a wide range of human behavior, including where phobias come from, why you dislike certain foods, the source of your emotions, how advertising works, why you feel anxi- ety before a job interview or an exam, and what arouses you sexually. Several later studies dealing with some of these applications are discussed here. Classical conditioning focuses on reflexive behavior: those behaviors that are not under your voluntary control. Any reflex can be conditioned to occur to a previously neutral stimulus. You can be classically conditioned so that your left eye blinks when you hear a doorbell, your heart rate increases at the sight of a flashing blue light, or you experience sexual arousal when you eat strawberries. The doorbell, blue light, and strawberries were all neutral in relation to the conditioned responses until they somehow became associated with uncondi- tioned stimuli for eye blinking (e.g., a puff of air into the eye), heart rate increase (e.g., a sudden loud noise), and sexual arousal (e.g., romantic caresses). To experience firsthand the process of classical conditioning, here is an experiment you can perform on yourself. All you will need is a bell, a mirror, and, to serve as your temporary laboratory, a room that becomes completely dark when the light is switched off. The pupils of your eyes dilate and constrict reflexively according to changes in light intensity. You have no voluntary control over this, and you did not have to learn how to do it. If I say to you “Please dilate your pupils now,” you would be unable to do so. However, when you walk into a dark theater, they dilate immediately. Therefore, a decrease in light would be considered an unconditioned stimulus for pupil dilation, the unconditioned response. In your temporary lab, ring the bell and, immediately after, turn off the light. Wait in the total darkness about 15 seconds and turn the light back on. Wait another 15 seconds and repeat the procedure: bell . . . light off . . . wait 15 seconds . . . light on . . . . Repeat this pairing of the neutral stimulus (the bell)

88 Chapter III Conditioning and Learning with the unconditioned stimulus (the darkness) 10 to 20 times, making sure that the bell only rings just prior to the sudden darkness. Now, with the lights on, watch your eyes closely in the mirror and ring the bell. You will see your pupils dilate slightly even though there is no change in light! The bell has become the conditioned stimulus and pupil dilation the conditioned response. relaTed research and recenT aPPlicaTions Two other studies presented in this book rest directly on Pavlov’s theory of classical conditioning. In the next article, John B. Watson conditioned 11-month-old Little Albert to fear a white rat (and other furry things) by employing the same principles Pavlov used to condition salivation in dogs. By doing so, Watson demonstrated how emotions, such as fear, are formed. Later, Joseph Wolpe (see Chapter IX: Psychotherapy: Reading 34) developed a therapeutic technique for treating intense fears (phobias) by applying the concepts of classical conditioning. His work was based on the idea that the association between the conditioned stimulus and the unconditioned stimu- lus must be broken in order to reduce the fearful response. This line of research on classical conditioning and phobias continues to the present. For example, studies have found that children whose parents have phobias may develop the same phobias to objects such as snakes and spiders through “vicarious” conditioning from mom and dad without any direct exposure to the feared object (Fredrikson, Annas, & Wik, 1997). The countless applications of Pavlov’s theory in the psychological and medical literature are far too numerous to summarize in any detail here. Instead, a few additional examples of the more notable findings are discussed. A common problem that plagues ranchers around the world is that of predatory animals, usually wolves and coyotes, killing and eating their livestock. In the early 1970s, studies were conducted that attempted to apply Pavlovian conditioning techniques to solve the problem of the killing of sheep by coyotes and wolves without the need for killing the predators (see Gustafson et al., 1974). Wolves and coyotes were given pieces of mutton (meat from sheep) containing small amounts of lithium chloride (UCS), a chemical that if ingested, makes an animal sick. When the animals ate the meat, they became dizzy, with severe nausea and vomiting (UCR). After recovering, these same hungry predators were placed in a pen with live sheep. The wolves and coyotes began to attack the sheep (CS), but as soon as they smelled their prey, they stopped and stayed as far away from the sheep as possible. When the gate to the pen was opened, the wolves and coyotes actually ran away from the sheep! Based on this and other related research, ranchers commonly use this method of classical conditioning to keep wolves and coyotes away from their herds. Another potentially vital area of research involving classical conditioning is in the field of behavioral medicine. Studies have suggested that the activity of the immune system can be altered using Pavlovian principles. Ader and Cohen (1985) gave mice water flavored with saccharine (mice love this water). They then paired the saccharine water with an injection of a drug that weakened the

Reading 9 It’s Not Just About Salivating Dogs! 89 immune system of the mice. Later, when these conditioned mice were given the saccharine water but no injection, they showed signs of immunosuppres- sion, a weakening of the immune response. Research is underway (primarily within a psychology subfield called psychoneuroimmunology) to study if the reverse is also possible—if immune enhancing responses may be classically con- ditioned. Overall, research is demonstrating that classical conditioning may indeed hold promise for increasing the effectiveness of immune system responses in humans (Miller & Cohen, 2001). Just imagine: In the future, you may be able to strengthen your resistance to illness by exposing yourself to a nonmedical conditioned stimulus. For example, imagine you feel the begin- nings of a cold or the flu, so you tune into your special classically conditioned “immune response enhancement music” on your iPod. As the music fills your ears, your resistance rises as a conditioned response to this stimulus and stops the disease in its tracks. As a demonstration of the continuing impact of Pavlov’s discoveries on today’s psychological research, consider the following. Since 2000, more than a thousand scientific articles have cited Pavlov’s work that forms the basis for this discussion. One especially fascinating recent study demonstrated how your psychological state at the time of conditioning and extinction may play a part in the treatment of classically conditioned irrational fears, called phobias (Mystkowski et al., 2003). Researchers used desensitization techniques to treat participants who were terrified of spiders. Some received the treat- ment after ingesting caffeine, while others ingested a placebo. A week later, all participants were retested—some receiving caffeine and others a placebo. Those who were given the placebo during treatment, but received real caf- feine at the follow-up, and those who had received real caffeine during treat- ment, but received a placebo at the follow-up, experienced a relapse of the fear response. In other words, changing the characteristics of a stimulus situ- ation lessens the effect of extinction. However, those who were in the same drug condition, either caffeine or placebo, at treatment and follow-up, con- tinued to experience a lowered fear response to spiders. This finding implies that if a classically conditioned behavior is successfully placed on extinction, the response may return, if the conditioned stimulus is encountered in a new and different situation. conclusion These examples demonstrate how extensive Pavlov’s influence has been on many scientific and research disciplines. For psychology in particular, few scientists have had as much impact in any single discipline. Classical conditioning is one of the fundamental theories on which modern psychology rests. Without Pavlov’s contributions, behavioral scientists still might have uncovered most of these principles over the decades. It is unlikely, however, that such a cohesive, elegant, and well-articulated theory of the conditioned reflex would ever have existed if Pavlov had not made the decision to risk his career and venture into the untested, uncharted, and highly questionable science of 19th-century psychology.

90 Chapter III Conditioning and Learning Ader, R., & Cohen, N. (1985). CNS-immune system interactions: Conditioning phenomena. Behavioral and Brain Sciences, 8, 379–394. Fredrikson, M., Annas, P., & Wik, G. (1997). Parental history, aversive exposure, and the develop- ment of snake and spider phobias in women. Behavior Research and Therapy, 35(1), 23–28. Gustafson, C. R., Garcia, J., Hawkins, W., & Rusiniak, K. (1974). Coyote predation control by aversive conditioning. Science, 184, 581–583. Miller, G., & Cohen, S. (2001). Psychological interventions and the immune system: A meta- analytic review and critique. Health Psychology, 20, 47–63. Mystkowski, J., Mineka, S., Vernon, L., & Zinbarg, R. (2003). Changes in caffeine states enhance return of fear in spider phobia. Journal of Consulting and Clinical Psychology, 71, 243–250. Reading 10: lIttle emotIoNAl AlbeRt Watson, J. B., & Rayner, R. (1920). Conditioned emotional responses. Journal of Experimental Psychology, 3, 1–14. Have you ever wondered where your emotions come from? If you have, you’re not alone. The source of our emotions has fascinated behavioral scientists throughout psychology’s history. Part of the evidence for this fascination can be found in this book; four studies are included that relate directly to emo- tional responses (Chapter V, Harlow, 1958; Chapter VI, Ekman & Friesen, 1971; Chapter VIII, Seligman & Meier, 1967; and Chapter IX, Wolpe, 1961). This study by Watson and Rayner on conditioned emotional responses was a strikingly powerful piece of research when it was published nearly a century ago, and it continues to exert influence today. You would be hard pressed to pick up a textbook on general psychology or on learning and behavior without finding a summary of the study’s findings. The historical importance of this study is not solely due to the research findings but also to the new psychological territory it pioneered. If we could be transported back to the turn of the century and get a feel for the state of psychology at the time, we would find it nearly completely dominated by the work of Sigmund Freud (see the reading on Anna Freud in Chapter VIII). Freud’s psychoanalytic view of human behavior was based on the idea that we are motivated by unconscious instincts and repressed conflicts from early childhood. In simplified Freudian terms, behavior, thoughts, and emotions are generated internally through biological and instinctual processes. In the 1920s, a new movement in psychology known as behaviorism, spearheaded by Pavlov (as discussed in the previous study) and Watson, began to take hold. The behaviorists’ viewpoint was radically opposed to the psychoanalytic school and proposed that behavior is generated outside the person through vari- ous environmental or situational stimuli. Therefore, Watson theorized, emotional responses exist in us because we have been conditioned to respond emotionally to certain stimuli that we encounter. In other words, we learn our emotional reac- tions. Watson (1913) believed that all human behavior was a product of learning and conditioning, as he proclaimed in his famous statement, Give me a dozen healthy infants, well-formed, and my own special world to bring them up in, and I’ll guarantee to take any one at random and train him to

Reading 10 Little Emotional Albert 91 become any type of specialist I might select—doctor, lawyer, artist, merchant- chief, and, yes, beggar man and thief. This was, for its time, an extremely revolutionary view. Most psychologists, as well as public opinion in general, were not ready to accept these new ideas. This was especially true for emotional reactions, which seemed to be gener- ated from within the person. Watson set out to demonstrate that specific emo- tions could be conditioned without regard for any internal forces. TheoreTical ProPosiTions Watson theorized that if a stimulus automatically produces a certain emotion in you (such as fear) and that stimulus is repeatedly experienced at the same moment as something else, such as a rat, the rat will become associated in your brain with the fear. In other words, you will eventually become conditioned to be afraid of the rat (this view reflects Pavlov’s theory of classical conditioning). He maintained that we are not born to fear rats but that such fears are learned through conditioning. This formed the theoretical basis for his most famous experiment, which involved a participant named “Little Albert.” MeThod and resulTs The participant, “Albert B.,” was recruited for this study at the age of 9 months from a hospital where he had been raised as an orphan from birth. The researchers and the hospital staff judged him to be very healthy, both emotionally and physically. To see if Albert was naturally afraid of certain stim- uli, the researchers presented him with a white rat, a rabbit, a monkey, a dog, masks with and without hair, and white cotton wool. Albert’s reactions to these stimuli were closely observed. Albert was interested in the various animals and objects and would reach for them and sometimes touch them, but he never showed the slightest fear of them. Because they produced no fear, these are referred to as neutral stimuli. The next phase of the experiment involved determining if a fear reaction could be produced by exposing Albert to a loud noise. This was not difficult, because all humans, and especially infants, will exhibit fear reactions to loud, sudden noises. Because no learning is necessary for this response to occur, the loud noise is called an unconditioned stimulus. In this study, a steel bar 4 feet in length was struck with a hammer just behind Albert. This noise startled and frightened him and made him cry. Now the stage was set for testing the idea that the emotion of fear could be conditioned in Albert. The actual conditioning tests were not done until the child was 11 months old. The researchers were hesitant to create fear reactions in a child experimentally, but they made the decision to proceed based on what was, in retrospect, questionable ethical reasoning. (This is discussed in con- junction with the overall ethical problems of this study, later in this review.) As the experiment began, Watson and his graduate research assistant, Rosalie Rayner, presented Albert with the white rat. At first, Albert was

92 Chapter III Conditioning and Learning interested in the rat and reached out to touch it. As he did this, the metal bar was struck, which startled and frightened Albert. This process was repeated three times. One week later, the same procedure was followed. After a total of seven pairings of the noise and the rat, the rat was presented to Albert alone, without the noise. As you’ve probably guessed by now, Albert reacted with extreme fear to the rat. He began to cry, turned away, rolled over on one side away from the rat, and began to crawl away so fast that the researchers had to rush to catch him before he crawled off the edge of the table! A fear response had been conditioned to an object that had not been feared only 1 week earlier. The researchers then wanted to determine if this learned fear would transfer to other objects. In psychological terms, this transfer is referred to as generalization. If Albert showed fear of other similar objects, then the learned behavior is said to have generalized. The next week, Albert was tested again and was still found to be afraid of the rat. Then, to test for generalization, an object similar to the rat (a white rabbit) was presented to Albert. In the author’s words, Negative responses began at once. He leaned as far away from the animal as possible, whimpered, then burst into tears. When the rabbit was placed in contact with him, he buried his face in the mattress, then got up on all fours and crawled away, crying as he went. (p. 6) Remember, Albert was not afraid of the rabbit prior to conditioning, and had not been conditioned to fear the rabbit specifically. Little Albert was presented over the course of this day of testing with a dog, a white fur coat, a package of cotton, and Watson’s own head of gray hair. He reacted to all of these items with fear. One of the most well-known tests of generalization that made this research as infamous as it is famous occurred when Watson presented Albert with a Santa Claus mask. The reaction? Yes . . . fear! After another 5 days Albert was tested again. The sequence of presenta- tions on this day is summarized in Table 10-1. Another aspect of conditioned emotional responses Watson wanted to explore was whether the learned emotion would transfer from one situation to another. If Albert’s fear responses to these various animals and objects occurred only in the experimental setting and nowhere else, the significance of the findings would be greatly reduced. To test this, later on the day outlined in Table 10-1, Albert was taken to an entirely different room with brighter lighting and more people present. In this new setting, Albert’s reactions to the rat and rabbit were still clearly fearful, although somewhat less intense. The final test that Watson and Rayner wanted to make was to see if Albert’s newly learned emotional responses would persist over time. Albert had been adopted and was scheduled to leave the hospital in the near future. Therefore, all testing was discontinued for a period of 31 days. At the end of this time, he was once again presented with the Santa Claus mask, the white fur coat, the rat, the rabbit, and the dog. After a month, Albert remained very afraid of all these objects.

Reading 10 Little Emotional Albert 93 Table 10-1 Sequence of Stimulus Presentations to albert on Fourth Day of Testing stImulus PReseNteD ReActIoN obseRveD 1. blocks Played with blocks as usual 2. Rat Fearful withdrawal (no crying) 3. Rat + Noise Fear and crying 4. Rat Fear and crying 5. Rat Fear, crying, and crawling away 6. Rabbit Fear, but less strong reaction than on former presentations 7. blocks Played as usual 8. Rabbit same as 6 9. Rabbit same as 6 10. Rabbit some fear, but also wanted to touch rabbit 11. Dog Fearful avoidance 12. Dog + Noise Fear and crawling away 13. blocks Normal play Watson and his colleagues had planned to attempt to recondition Little Albert and eliminate these fearful reactions. However, Albert left the hospital on the day these last tests were made, and, as far as anyone knows, no recondi- tioning ever took place. discussion and significance of findings Watson had two fundamental goals in this study and in all his work: (a) to demonstrate that all human behavior stems from learning and conditioning and (b) to demonstrate that the Freudian conception of human nature, that our behavior stems from unconscious processes, was wrong. This study, with all its methodological flaws and serious breaches of ethical conduct, suc- ceeded to a large extent in convincing many in the psychological community that emotional behavior could be conditioned through simple stimulus- response techniques. This finding helped, in turn, to launch one of the major schools of thought in psychology: behaviorism. Here, something as complex and personal as an emotion was shown to be subject to conditioning, just as Pavlov demonstrated that dogs learn to salivate at the sound of a metronome. A logical extension of this is that other emotions, such as anger, joy, sad- ness, surprise, or disgust, may be learned in the same manner. In other words, the reason you are sad when you hear that old song, nervous when you have a job interview or a public speaking engagement, happy when spring arrives, or afraid when you hear a dental drill is that you have developed an association in your brain between these stimuli and specific emotions through conditioning. Other more extreme emotional responses, such as phobias and sexual fetishes, may also develop through similar sequences of conditioning. Watson was quick to point out that his findings could explain human behavior in rather straightforward and simple terms, compared with the complexities of the psychoanalytic notions of Freud and his followers.

94 Chapter III Conditioning and Learning As Watson and Rayner explained in their article, a Freudian would explain thumb sucking as an expression of the original pleasure-seeking instinct. Albert, however, would suck his thumb whenever he felt afraid. As soon as his thumb entered his mouth, his fear lessened. Therefore, Watson interpreted thumb sucking as a conditioned device for blocking fear-producing stimuli. An additional questioning of Freudian thinking in this article concerned how Freudians in Albert’s future, given the opportunity, might analyze Albert’s fear of a white fur coat. Watson and Rayner claimed that Freudian analysts “will probably tease from him the recital of a dream which, upon their analy- sis, will show that Albert at three years of age attempted to play with the pubic hair of the mother and was scolded violently for it” (p. 14). Their main point was that they had demonstrated with Little Albert that emotional disturbances in adults cannot always be attributed to sexual traumas in childhood, as the Freudian view maintained. QuesTions and criTicisMs As you have been reading this, you have probably been concerned or even angered over the experimenter’s treatment of this innocent child. This study clearly violated current standards of ethical conduct in research involving humans. It would be highly unlikely that any institutional review board at any research institution would approve this study today. A century ago, however, such ethical standards did not formally exist, and it is not unusual to find reports in the early psychological literature of what now appear to be questionable research methods. It must be pointed out that Watson and his colleagues were not sadistic or cruel people and that they were engaged in a new, unexplored area of research. They acknowledged their considerable hes- itation in proceeding with the conditioning process but decided that it was justifiable, because, in their opinion, some such fears would arise anyway when Albert left the sheltered hospital environment. Even so, is it ever appropriate to frighten a child to this extent, regardless of the importance of the potential discovery? Today nearly all behavioral scientists would agree that it is not. Another important point regarding the ethics of this study was the fact that Albert was allowed to leave the research setting and was never recondi- tioned to remove his fears. Watson and Rayner contended in their article that such emotional conditioning may persist over a person’s lifetime. If they were correct on this point, it is extremely difficult, from an ethical perspective, to justify allowing someone to grow into adulthood fearful of all these objects (and who knows how many others!). Several researchers have criticized Watson’s assumption that these con- ditioned fears would persist indefinitely (e.g., Harris, 1979). Others claim that Albert was not conditioned as effectively as the authors maintained (e.g., Samelson, 1980). Researchers have frequently demonstrated that behaviors acquired through conditioning can be lost because of other experiences or simply because of the passage of time. Imagine, for example, that when Albert turned age five, he was given a pet white rabbit for a birthday present.

Reading 10 Little Emotional Albert 95 At first, he might have been afraid of it (no doubt baffling his adoptive par- ents). As he continued to be exposed to the rabbit without anything frighten- ing occurring (such as that loud noise), he would probably slowly become less and less afraid until the rabbit no longer caused a fear response. This is a well- established process in learning psychology called extinction, and it happens routinely as part of the constant learning and unlearning, conditioning and unconditioning processes we experience throughout our lives. However, it appears this was not true for Albert. Recent research by a team of graduate students that took over 7 years has finally shed light on who the child called “Albert” truly was: a mystery that has persisted for decades. “Albert has been identified as Douglas Merritte, the son of Arvilla Merritte, a nurse at the Johns Hopkins University Hospital department called the Harriet Lane Home during the years when Watson’s studies were carried out. She was paid $1.00 for her permission for her son to be studied (see DeAngelis, 2010). The saddest part of these newly discovered aspects of the study was that Douglas died at the age of 6 from hydrocephalus, a condition involving a build up of fluid in and around the brain. Today, many treatments exist for this serious condition, but in the 1920s it was often fatal. Consequently, the question of whether “Albert” was ever able to overcome the fears that Watson had conditioned in him, remains unknown. Furthermore, as long as we are in the “criticisms” section of this reading, Watson was dis- missed from the Johns Hopkins University around the time of this study for engaging in an affair with his graduate assistant, Rosalie Rayner. recenT aPPlicaTions Watson’s 1920 article, its ethical flaws notwithstanding, continues to be cited in research in a wide range of applications, including theories of effective parenting and psychotherapy. One study, examined the facial expressions of emotion in infants (Sullivan & Lewis, 2003). We know that facial expressions corresponding to specific emotions are consistent among all adults and across cultures (see the reading on Ekman’s research in Chapter VI). This study, how- ever, extended this research to how such expressions develop in infants and what the various expressions mean at very young ages. A greater understanding of infants’ facial expressions might be of great help in adults’ efforts to com- municate with and care for babies. The authors noted that their goal in their research was “to provide practitioners with basic information to help them and the parents they serve become better able to recognize the expressive signals of the infants and young children in their care” (p. 120). These authors’ use of Watson’s findings offers us a degree of comfort in that his questionable research tactics with Little Albert, may, in the final analysis, allow us to develop greater sensitivity and perception into the feelings and needs of infants. As mentioned previously in this discussion, one emotion—fear—in its extreme form, can produce serious negative consequences known as phobias. Many psychologists believe that phobias are conditioned much like Little Albert’s

96 Chapter III Conditioning and Learning fear of furry animals (see the discussion of Wolpe’s research on the treatment of phobias in Chapter IX: Psychotherapy). Watson’s research has been incorpo- rated into many studies about the origins and treatments of phobias. One such article discussed phobias from the nature–nurture perspective and found some remarkable results. Watson’s approach, of course, is rooted completely in the environmental or nurture side of the argument, and most people would view phobias as learned. However, a study by Kendler, Karkowski, and Prescott (1999) provided compelling evidence that the development of phobias may include a substan- tial genetic component. The researchers studied phobias and unreasonable fears in more than 1,700 female twins (see the discussion of Bouchard’s twin research in Chapter I). They claim to have found that a large percentage of the variation in phobias was due to inherited factors. The authors concluded that, although phobias may be molded by an individual’s personal experiences, the role of a person’s family in the development of phobias is primarily genetic, not environmental. Imagine: Born to be phobic! This view flies directly in the face of Watson’s theory and should provide plenty of fuel for the ongoing nature– nurture debate in psychology and throughout the behavioral sciences. DeAngelis, T. (2010). Little Albert regains his identity. Monitor on Psychology, 41, 10. Harris, B. (1979). What ever happened to Little Albert? American Psychologist, 34, 151–160. Kendler, K., Karkowski, L., & Prescott, C. (1999). Fears and phobias: reliability and heritability. Psychological Medicine, 29(3), 539–553. Samelson, F. (1980). Watson’s Little Albert, Cyril Burt’s twins, and the need for a critical science. American Psychologist, 35, 619–625. Sullivan, M., & Lewis, M. (2003). Emotional expressions of young infants and children: A practi- tioner’s primer. Infants and Young Children, 16, 120–142. Watson, J. B. (1913). Psychology as the behaviorist views it. Psychological Review, 20, 158–177. Reading 11: KNoCK WooD! Skinner, B. F. (1948). Superstition in the pigeon. Journal of Experimental Psychology, 38, 168–172. In this reading, we examine one study from a huge body of research carried out by one of the most influential and most widely known figures in the history of psychology: B. F. Skinner (1904–1990). Deciding how to present Skinner and which of his multitude of studies to explore is a difficult task. It is impos- sible to represent adequately in one short article Skinner’s contributions to the history of psychology. After all, Skinner is considered by most to be the father of radical behaviorism, he was the inventor of the famous (or infamous) Skinner Box, and he was the author of over 20 books and many hundreds of scientific articles. This article, with the funny-sounding title “Superstition in the Pigeon,” has been selected from all his work because it allows for a clear discussion of Skinner’s basic theories, provides an interesting example of his approach to studying behavior, and offers a “Skinnerian” explanation of a behavior with which we are all familiar: superstition.

Reading 11 Knock Wood! 97 Skinner is referred to as a radical behaviorist because he believed that all behaviors—including public (or external) behavior, as well as private (or internal) events such as feelings and thoughts—are ultimately learned and controlled by the relationships between the situation that immediately precedes the behavior and the consequences that directly follow it. Although he believed that private behaviors are difficult to study, he acknowledged that we all have our own subjective experience of these behaviors. He did not, however, view internal events, such as thoughts and emotions, as causes of behavior but rather as part of the mix of environment and behavior that he was seeking to explain (see Schneider & Morris, 1987, for a detailed discussion of the term radical behaviorism). To put Skinner’s theory in very basic terms, in any given situation, your behavior is likely to be followed by consequences. Some of these consequences, such as praise, receiving money, or the satisfaction of solving a problem, will make the behavior more likely to be repeated in future, similar situations. These consequences are called reinforcers. Other consequences, such as injur- ing yourself or feeling embarrassed, will tend to make the behavior less likely to be repeated in similar situations. These consequences are called punishers. The effects of these relationships between behavior and the environment are called reinforcement and punishment respectively (Edward  K. Morris, per- sonal communication, September 1987). Reinforcement and punishment are two of the most fundamental processes in what Skinner referred to as operant conditioning and may be diagrammed as follows: Situation Behavior Consequence Reinforcement = Learning Punishment = No learning Within this conceptualization, Skinner also was able to explain how learned behaviors decrease and sometimes disappear entirely. When a behav- ior has been reinforced and the reinforcement is then withdrawn, the likeli- hood of the behavior reoccurring will slowly decrease until the behavior is effectively suppressed. This process of behavior suppression is called extinction. If you think about it, these ideas are not new to you. The process we use to train our pets follows these same rules. You tell a dog to sit, it sits, and you reward it with a treat. After a while the dog will sit when told to, even without an immediate reward. You have applied the principles of operant condition- ing. This is a very powerful form of learning and is effective with all animals, even old dogs learning new tricks and, yes, even cats! Also, if you want a pet to stop doing something, all you have to do for the behavior to stop is remove the reinforcement. For example, if your dog is begging at the dinner table, there is a reason for that (regardless of what you may think, dogs are not born to beg at the table). You have conditioned this behavior in your dog through reinforcement. If you want to put that behavior on extinction, the reinforcement must be totally discontinued. Eventually, the dog will stop begging. By the way,

98 Chapter III Conditioning and Learning if one member of the family cheats during extinction and secretly gives the canine beggar some food once in a while, extinction will never happen, but the dog will spend much more of its begging energy near that person’s chair. Beyond these fundamentals of learning, Skinner maintained that all human behavior is created and maintained in precisely the same way. It’s just that with humans, the exact behaviors and consequences are not always easy to iden- tify. Skinner was well-known for arguing that if a human behavior was interpreted by other theoretical approaches to be due to our highly evolved consciousness or intellectual capabilities, it was only because those theorists had been unable to pinpoint the reinforcers that had created and were maintaining the behavior. If this feels like a rather extreme position to you, remember that Skinner’s position was called radical behaviorism and was always surrounded by controversy. Skinner often met skepticism and defended his views by demonstrating experimentally that behaviors considered to be the sole property of humans could be learned by “lowly creatures” such as pigeons or rats. One of these demonstrations involved the contention by others that superstitious behavior is uniquely human. The argument was that superstition requires human cognitive activity (i.e., thinking, knowing, reasoning). A superstition is a belief in something, and we do not usually attribute such beliefs to animals. Skinner said in essence that superstitious behavior could be explained as easily as any other action by using the principles of operant conditioning. He performed this experiment to prove it. TheoreTical ProPosiTions Think back to a time when you have behaved superstitiously. Did you knock on wood, avoid walking under a ladder, avoid stepping on cracks, carry a lucky coin or other charm, shake the dice a certain way in a board game, or change your behavior because of your horoscope? It is probably safe to say that everyone has done something superstitious at some time, even if some of them might not want to admit it. Skinner said that the reason people do this is that they believe or presume a connection exists between the superstitious behavior in a certain setting and a reinforcing consequence, even though—in reality—it does not. This connection exists because the behavior (such as shaking the dice that certain way) was accidentally reinforced (by something rewarding, such as a good roll) once, twice, or several times. Skinner called this noncontingent reinforcement—that is, a reward that is not contingent on any particular behav- ior. You believe that there is a causal relationship between the behavior and the reward, when no such relationship exists. “If you think this is some exclusive human activity,” Skinner might have said, “I’ll create a superstitious pigeon!” MeThod To understand the method used in this experiment, a brief description of what has become known as the Skinner Box is necessary. The principle behind the Skinner Box (or conditioning chamber, as Skinner called it) is really quite simple.

Reading 11 Knock Wood! 99 It consists of a cage or box that is empty except for a dish or tray into which food may be dispensed. This allows a researcher to have control over when the animal receives reinforcement, such as pellets of food. The early conditioning boxes also contained a lever, which, if pressed, would cause some food to be dispensed. If a rat (rats were used in Skinner’s earliest work) was placed in one of these boxes, it would eventually, through trial and error and reinforcement, learn to press the lever for food. Alternatively, the experimenter could, if desired, take control of the food dispenser and reinforce a specific behavior. Later, Skinner and others found that pigeons also made ideal subjects in con- ditioning experiments, and conditioning chambers were designed with disks to be pecked instead of bars to be pressed. These conditioning cages were used in the study discussed here, but with one important change. To study superstitious behavior, the food dispensers were rigged to drop food pellets into the tray at intervals of 15 seconds, regardless of what the animal was doing at the time. The reward was not contin- gent on any particular behavior. This was noncontingent reinforcement: The animal received a reward every 15 seconds, no matter what it did. Subjects in this study were eight pigeons. These birds were fed less than their normal daily amount for several days so that when tested they would be hungry and therefore motivated to perform behaviors for food (this increased the power of the reinforcement). Each pigeon was placed into the experimen- tal cage for a few minutes each day and just left there to do whatever a pigeon does. During this time, reinforcement was being delivered automatically every 15 seconds. After several days of conditioning in this way, two independent observers recorded the birds’ behavior in the cage. resulTs As Skinner reports, In six out of eight cases the resulting responses were so clearly defined that two observers could agree perfectly in counting instances. One bird was conditioned to turn counterclockwise about the cage, making two or three turns between reinforcements. Another repeatedly thrust its head into one of the upper corners of the cage. A third developed a tossing response as if placing its head beneath an invisible bar and lifting it repeatedly. Two birds developed a pendulum motion of the head and body in which the head was extended forward and swung from right to left with a sharp movement followed by a somewhat slower return. The body generally followed the movement and a few steps might be taken when it was extensive. Another bird was conditioned to make incomplete pecking or brushing movements directed toward but not touching the floor. (p. 168) None of these behaviors had been observed in the birds prior to the condi- tioning procedure. The new behaviors had no real effect on the delivery of food. Nevertheless, the pigeons behaved as if a certain action would produce the food—that is, they became superstitious. Skinner next wanted to see what would happen if the time interval between reinforcements was extended. With one of the head-bobbing and