What is Learning?
Module 01 Reading
Chapter 1
What is Learning?
DEFINING LEARNING
Evolutionary Adaptation
Adaptive Behavior
Learning Definition
Important Distinctions of the Definition
Learning versus Performance
Temporary Changes Don’t Count
Example of the Definition in Action
Check Your Learning: Defining Learning
BIOLOGICAL CONSTRAINTS ON LEARNING
Prepared Behaviors
Unprepared Behaviors
Contraprepared Behaviors
Box 1.1: The Significance of Animal Research
Check Your Learning: Biological Constraints
A 5-year-old can recite the alphabet because of singing the “alphabet song” at preschool.
A cat comes running when it hears the electric can opener running because its owner uses the can opener to open cans of cat food.
A college student understands that certain songs evoke certain memories because of the principles of classical conditioning.
A chimp uses a stick as a tool to knock down bananas because it has seen other chimps do so.
A sixth-grade boy is fearful of dogs because he was attacked by a dog when he was 3 years old.
Despite a wide variety of species, ages, and behaviors portrayed in these scenarios, they all have a common element.
_________________________________________________________________
Think Ahead
**** Review the scenarios listed above. What do you think all these situations share in common? Before reading further, take a moment to write down your thoughts.
________________________________________________________________
All of these scenarios illustrate some form of learning, which is an important adaptive process used by humans and other animals to respond to stimuli in their environments. It is imperative that organisms adapt to their environments in order to ensure their survival.
DEFINING LEARNING
We have already used the word “learning,” and you have no doubt used the word thousands of times in your life. Yet, from a scientific perspective, we need to step back and ask what the concept of learning actually means – to formulate an operational definition of learning.
Adaptive Behavior
Perhaps the first characteristic of learning we can agree on is that it is an adaptive behavior. Adaptation can take one of two forms: evolutionary or learning. In evolutionary adaptation, also known as natural selection, certain biological traits change over time to help the species (or subset of the species) survive. One of the best-known examples of evolutionary adaptation is the story of the peppered moth in England (see Figure 1.1). The peppered moth exists in two colors – a darker and lighter version. Prior to the Industrial Revolution in England, the lighter colored moth predominated – its coloring was such that it was difficult to spot among the lichens growing on the trees. Thus, it was less of a target for predatory birds. However, pollutants released during industrialization killed the lichens on the trees, with the result that the tree’s natural darker bark was exposed. Because of this change, the darker moths were less visible to birds than the lighter moths. The predominance of colored moths then switched, as darker moths became more common. The adaptive process of natural selection operated to make darker moths more likely to survive and reproduce, whereas the lighter moths were less likely to survive and reproduce. This evolutionary adaptation accounted for the change from approximately 1% of the moths being dark in 1848 to about 90% in 1959. Interestingly, after England passed laws to outlaw the pollution from industrialization, the lichen grew back on trees and, over time, the lighter moths have again become more common.
[[Insert Figure 1.1., photo of pepper moths, about here]]
Other common examples of evolutionary adaptation are insects that become resistant to pesticides over time and a current potential crisis in medicine – the ability of bacteria and viruses to change so that they are resistant to antibiotics that have been used to treat them in the past. Both examples showcase natural selection – the resistant organisms are the ones that survive and reproduce. This evolutionary adaptation refers to adapting some physical characteristic that promotes survival.
In contrast, the type of adaptation that we are interested in examining in this textbook, learning, refers to adaptive behavior – not merely an adaptation in a physical characteristic. Learning may promote survival, like evolutionary adaptation does, or it may be useful in some other way. Learning is an adaptive process because the new behaviors an organism learns and can emit may enable it to meet environmental challenges that it could not have previously met. As a result, an organism that learns is more likely to survive.
Learning Definition
Learning is a broad term that encompasses many processes we will cover in the coming chapters. In addition, it is related to many important psychological concepts that are, themselves, major topics of research and study. Hillner (1978) presented an illuminating list of phenomena that psychologists might study – the breadth of this list (see Table 1.1) will give you some idea of the difficulty involved in defining learning. Navarick (1979) noted that examples of learning are easy to come by, but that there was no universally accepted definition of the term.
[[Insert Table 1.1, Hillner’s (1978) List . . . , about here]]
Because learning is somewhat difficult to define, many experts have developed differing definitions for the term. Two of the more influential definitions came from Kimble (1967) and Hilgard and Bower (1975). According to Kimble (p. 82), “learning is a relatively permanent change in a behavior potentiality which occurs as a result of reinforced practice.” Hilgard and Bower, on the other hand, used the following definition:
Learning refers to the change in a subject’s behavior to a given situation brought about by his repeated experiences in that situation, provided that the behavior change cannot be explained on the basis of native response tendencies, maturation, or temporary states of the subject (e.g., fatigue, drugs, etc.). (p. 17)
A comparison of these two definitions reveals several differences: a relatively permanent change versus a change, behavior potentiality versus behavior, reinforced practice versus repeated experiences, several exceptions versus none.
Consideration of these different definitions plus examining relevant research led Tarpy and Mayer (1978) to propose the following definition: “Learning is an inferred change in the organism’s mental state which results from experience and which influences in a relatively permanent fashion the organism’s potential for subsequent adaptive behavior” (p. 3). This is the definition that we will adopt (with minor grammatical changes) for use in this text.
Given the difficulty involved in defining learning, it should come as no surprise that the definition is a long one with several important nuances. Let’s take a look at the various important components of this definition.
Learning Is an Inferred Change. Learning is what is known as a hypothetical construct, something shared by many concepts in psychology. As you will see in Chapter 12, a hypothetical construct is a variable that we believe exists and that we use to explain events but that we cannot directly observe. If you look back at the five behaviors described at the beginning of this chapter, we use the variable of “learning” to explain why all five have occurred. We cannot see learning directly – instead, we see the outcome or byproduct of learning – usually a behavior of some sort – and then infer that learning has taken place.
It would be nice if we could measure learning directly, but we cannot. Instead, we decide that learning has taken place when an organism can perform some behavior that it previously could not perform. Until the organism performs the behavior in question, we do not know that learning has taken place. For example, suppose that you memorize this definition of learning we are using. You have learned the definition, and even if you do not write it on a test or tell it to someone, you have the ability to use that knowledge if the proper situation arises. If someone asks you the definition of learning, you can tell them the word-for-word definition that you learned.
Learning is an Inferred Change in the Organism’s Mental State. Although psychologists and neuroscientists are much closer to discovering how learning takes place within an organism, we are still not certain exactly what change occurs. Still, we are virtually certain that some physical change takes place in the nervous system. One strong candidate for explaining learning is long-term potentiation (LTP). LTP occurs when two neurons (or a chain of neurons) have been stimulated together several times. A change occurs at the synaptic level of the neurons and makes future communication between the pair easier. LTP is a popular explanation for learning and memory. The idea is that the perception of stimuli causes neurons to fire, so that the same chain of neurons might fire whenever you see a car (or any other stimulus). Because that same chain of neurons is stimulated, you have the memory of “car,” and you can identify a car when you see one.
Suppose you buy a new red car: Although you have a memory for car, you do not yet have the memory for “my new red car.” Instead, you may have a memory for “my old blue car.” After you see and drive your new car several times, you develop a memory for “my red car.” As you activate this memory more and more times, you remember to look for “my red car” in the parking lot instead of “my previous blue car” that you used to drive.
Learning Involves an Inferred Change in the Organism’s Mental State that Results from Experience. Learning does not occur by magic, divine inspiration, or in the absence of experience. Instead, for learning to occur, the organism must undergo “a particular instance of personally encountering or undergoing something” (dictionary.com). In the previous example, you had to personally encounter your new red car several times to learn that new car. When you were younger, you had to personally encounter and interact with a computer to learn how to use it. You did not simply wake up one morning and magically know how to use a computer.
In this definition, “experience” is a broad term that can encompass many different types of encountering and undergoing something. That wide variety does not mean that the experiences that can lead to learning are vague; rather, the wide variety means that there are many different types of experiences that can lead to learning.
Learning is an Inferred Change in the Organism’s Mental State that Results from Experience and that Influences in a Relatively Permanent Fashion. One major question about learning revolves around its permanence. Everyone has had the experience of forgetting something that they have learned, so the notion of learning being relatively permanent may come as a surprise to you. Of course, the hedge word “relatively” means that forgetting is possible; still, even “relatively permanent” may seem to be an overstatement to you if you are the forgetful type.
For many years, psychologists have tried to determine what happens to a memory when we forget it. Although the popular notion in the general public is that a forgotten memory is totally gone or erased, research evidence indicates otherwise. You can probably remember when something you thought you had forgotten suddenly “popped” into your mind – for example, your second grade teacher’s name, your old telephone number, or the time you ate too much cake and ice cream at your birthday party and got sick. The fact that “forgotten” memories can suddenly reappear supports the notion that memories may be stored permanently (or nearly so), but that we can lose the ability to retrieve them (pull them back out of memory). So, we may not have actually forgotten some specific information we want, but we may have forgotten how to get to that information. This situation is not unlike a file that you have stored on your computer but for which you have forgotten the file’s name. The file is still there, but you simply can’t access it.
Learning is an Inferred Change in the Organism’s Mental State that Results from Experience and that Influences in a Relatively Permanent Fashion the Organism’s Potential for Subsequent Adaptive Behavior. As we mentioned earlier in this chapter, learned behaviors are adaptive for survival or some other reason. Although it is pretty unlikely that knowing the definition of learning will aid your survival, it will help you understand the concept and (we hope) pass the class in which you are enrolled.
Likewise, the vast array of learned behaviors that you exhibit is useful for a wide variety of reasons. What those learned behaviors have in common is that they allow you to adapt to various situations you encounter. Knowing how to print and write were adaptive to your early school years, knowing how to type has been adaptive in your later school years, knowing how to write (grammar and composition) has been adaptive in your college career, knowing how to deal with bosses will be adaptive in your work career, and so on.
Behaviors that are not adaptive might be learned, but often will disappear over time. For example, as a child, you learned to cry when you wanted food or a diaper change – however, crying at this point in your life is not an effective way of getting something to eat, so that behavior is no longer adaptive for you. Thus, over time, crying for food has disappeared from your behavioral repertoire.
With these details in mind, we can understand how to distinguish examples of learning from changes in behavior that do not represent instances of learning. It is also important to remember how learning takes place—as a student, as a future parent, and as a member of society who draws conclusions about people’s behaviors.
Important Distinctions of the Definition
Because we adopt this definition of learning, we must make some distinctions of terms that have some important implications for our study of learning.
Learning versus Performance. As we mentioned previously, it is impossible to know whether learning has occurred by directly observing it. It would be nice if this truth were not so – if we did have some direct measure of learning. One idea that you may have heard as a child is that the brain forms a new wrinkle every time we learn something. Of course, this idea is not literally true: We are not born with smooth brains that wrinkle as we learn. If they did, we would conveniently be able to see if learning had occurred by observing wrinkling of the brain. Instead, however, we must observe learning indirectly by looking at the performance of behaviors. Someone knows that you learned how to write by observing you actually writing. Your professor will know that you have learned the definition of learning that we are using by asking you a question on an exam about that definition.
Tolman and Honzik (1930) conducted a classic experiment that clearly demonstrates the difference between learning and performance. To understand their experimental design, we first need to introduce the concept of reinforcement. Reinforcement is defined as a consequence that follows a behavior that strengthens that behavior and increases its future probability. Tolman and Honzik had three groups of rats running in a maze for 17 days. One group (HR) consistently received reinforcement (in this case, food) when it reached the end of the maze. A second group (HNR) consistently never found food at the end of the maze. The third group (HNR-R) did not find food at the end of the maze for the first 10 days; however, beginning on the 11th day, this group consistently received food at the end of the maze. Looking at Figure 1.2, you can see that the rats that always received food at the end of the maze improved their performance (made fewer errors) over the 17 days. The rats that never found food at the end of the maze also improved their performance, but not nearly as much, or as quickly, as the reinforced rats did. This finding is a pretty standard set of results: Organisms tend to learn more efficiently when they receive reinforcement for correct responses. (We will talk more about the effects of reinforcement in later chapters.) The third group was the critical group – these rats showed performance that was a combination of the other two groups. For the first 11 days, their performance was essentially the same as the rats that did not receive reinforcement. However, once they found reinforcement at the end of the maze, these rats very quickly improved their maze performance to equal or surpass the rats that had been reinforced from the beginning of the experiment.
[[Insert Figure 1.2, Results from Tolman . . . , about here]]
The term that learning psychologists give to this type of behavior is latent learning. The idea is that the third group of rats was learning the maze all along, but simply was not performing well on the maze because there was no reason to do so. Thus, you see the distinction between learning and performance in the data from the third group of rats.
One inconvenient implication of having to observe learning by looking at performance is that there are factors that can affect performance other than learning. Motivation is one such factor. Motivation, like learning, is an internal process. It serves to energize behavior, guides an organism’s behavior toward a goal, and maintains the behavior until te desired goal is achieved.
Suppose, for example, that you have trained your dog to shake hands by giving her a treat each time she shakes hands. What happens when you try to get your dog to shake hands but she is not hungry? If her motivation for shaking hands is the treat, she may not perform. An outside observer would likely conclude that your dog has not learned to shake hands because she does not perform the behavior. Apparently, the third group of rats in Tolman and Honzik’s (1930) experiment was not motivated to perform well until they found food at the end of the maze. Presumably, if Tolman and Honzik had introduced reinforcement to the nonreinforced rats on the 11th day (or even later), their performance would have quickly changed to match or surpass that of the consistently reinforced rats.
Other factors, such as emotions, can also reduce or interfere with performance. Emotion can be defined as physiological changes and conscious feelings of pleasantness or unpleasantness, aroused by external or internal stimuli, that lead to behavioral reactions. If you know anyone with young children, you have probably had the experience of seeing a proud parent who wants you to watch a child who has learned some new behavior. However, the parent is often disappointed when the child will not perform because of being shy or embarrassed. You may have wondered whether the child had actually learned the behavior. The parent insisted (correctly, of course) that the child had learned the behavior; however, the child did not perform the behavior because the emotion aroused interfered with the performance of the behavior.
An incentive is a factor related to motivation that can also affect performance. However, unlike motivation, which deals with a factor that is internal for the organism, incentives are external to the organism. Your dog is motivated to shake hands because she very much likes the dog treats you offer. Thus, the incentive value of the dog treats is high. If the incentive value of the treat is high enough, the dog will shake hands even if she is not hungry. However, if there is a food that your dog dislikes, she may not shake hands for that food, even if she is hungry. The incentive value of that food is low, so it overrides the motivation that the dog has for food. In a similar fashion, when you are driving on the highway and begin to feel hungry, you will bypass restaurants that serve food that you don’t like – instead, you will wait until you see your favorite fast-food restaurant. However, if you drive many miles looking for something you like, you may get so hungry that you stop at a restaurant that is not your favorite. In this situation, your motivation has overwhelmed the low incentive value of the food.
All of these factors affect performance rather than learning. Because we can only observe performance and not learning, these factors may mislead an observer into believing that an organism has not learned a certain behavior. However, if you think of an everyday example, you will easily see the fallacy in this conclusion. Imagine that you see a person sitting on a park bench. From simply looking at this individual, can you tell if he has learned to play the guitar, use a computer, or shoot a basketball? No, there is no way to tell! Suppose he has a guitar case, computer bag, or basketball sitting beside him on the bench. In this case, you might be tempted to conclude that he has learned the behavior appropriate to the object beside him, but you could not be certain about that conclusion. The only way to know for sure if he has learned the appropriate behavior is to observe him performing that behavior.
Temporary Changes Don’t Count. Suppose your dog gets very sick and will not shake hands in order to get a dog treat. Does this nonperformance mean that the dog has not learned the behavior? Does it mean that your dog has forgotten how to shake hands? The answer to both questions, of course, is clearly no. Likewise if you break your leg and have to use a wheelchair for a while, your inability to walk does not mean that you didn’t learn to walk or that you forgot how to walk.
The “relatively permanent” aspect of the definition precludes these temporary changes as being considered as either learning or unlearning. Any factor that affects learning for just a short amount of time should not be considered as part of the learning process – instead, we are seeing another example of learning versus performance.
The Hilgard and Bower (1975) definition of learning also excluded maturational and native (innate) responses. A maturational response is a behavior that becomes possible because an organism has become older, or developed. A native (innate) response is a behavior that an organism makes naturally without learning, what you might already know as an instinctive behavior. Our definition excludes these types of factors because they are not due to experience.
As an example of a maturational change that is not due to learning, consider a young child who has a toy basketball goal just three feet tall and a small sponge basketball. Although the child can learn to shoot baskets on the short goal, she cannot shoot baskets on a regulation 10-foot goal because she cannot throw the ball high enough. However, as she grows older and gets taller, she can eventually shoot baskets on the tall goal. This change is not due to her learning how to shoot the basketball; she has simply matured to the point that she is tall enough to shoot at a 10-foot basket.
Likewise, innate responses are not due to learning. If you pop a piece of hard candy in your mouth, you will salivate. Is this a learned response? No, salivation is an innate response to having food in your mouth. It occurs automatically to food stimuli to help in breaking food down in preparation for swallowing and digestion. Babies do not have to learn how to salivate – they naturally salivate from birth in response to having food in the mouth.
_______________________________________________________________
Think Ahead
******Can you come up with other examples of behaviors that people might interpret as an organism NOT having learned, but that are actually due to a motivational problem, emotion, and inadequate incentives? Try to think of one example of each. Next, can you think of examples in which a temporary situation, a maturational change, and an innate response influence behavior and interfere with the performance of a learned behavior? Before reading further, take a few minutes to write down your ideas.
________________________________________________________________
Of course, there are many possible answers to these questions. The key is that you understand the concepts and differences among them. The following answers are simply examples of each category; yours will probably not match ours, but they should have the same principles involved.
Motivational problem: You have watched your older sister make you a grilled cheese sandwich many times. However, you have never made a grilled cheese sandwich for yourself because she has always done it for you. One day, you want a grilled cheese sandwich, and your sister is gone. You are able to make the sandwich because you have watched her so many times. Because your dad had never seen you make a grilled cheese sandwich, he might believe that you had not learned that skill. However, you were simply not motivated to make a grilled cheese sandwich because you did not need to, thanks to your sister.
Emotion: You have learned your lines for your role in a school play. Everything goes well during daily rehearsals and the dress rehearsal. However, on the night of the performance, with a filled auditorium, you experience stage fright, and your mind goes blank, and you cannot say your lines. A member of the audience might be inclined to conclude that you had actually not learned your lines.
Inadequate incentive: You have just eaten a meal and would like something sweet to top it off. When you ask the waitress about dessert, she says they are out of everything except pumpkin pie. Your feeling about pumpkin pie is that it is akin to a vegetable, so you do not order dessert. Because the incentive (pumpkin pie) is not enticing to you, you do not exhibit the behavior of ordering dessert.
Temporary situation: Tiger Woods, the world’s best golfer, broke his leg in 2008, so he was unable to play golf. In 2009, when he started playing again, it would not be correct to say that he just learned to play golf. His inability to play was not related to a learning variable, but instead to a temporary situation.
Maturational change: This is probably the most difficult situation for an example because it is often difficult to distinguish maturation and learning. It is often the case that maturation has to take place before learning can occur. For example, young birds can’t fly until muscular and nervous system maturation take place. Once a bird has matured physically, flight seems to come naturally – almost innately (Biological Sciences Curriculum Studies, 2002). However, there is still a learned component to flying because baby birds do not fly as well as adult birds.
Innate response: You are impressed at how quickly your newborn child has learned to grasp your finger when you put it in her palm. However, later you are disappointed to find out that this behavior is simply known as the grasping reflex – something that all babies do innately. Further evidence of this behavior being a reflex is that it disappears after a few months of life – this disappearance is not forgetting, but simply a reflex going away after some time.
Example of the Definition in Action
Let’s take an in-depth look at a particular behavior to see how it fits within this definition. In the early 1900s, at least two species of bird (blue tit [titmouse in the U.S.], red robin) in England learned how to procure a tasty treat for themselves (de Geus, 1997). During this era, milk was delivered to the doorstep of homes – the milk was in a glass bottle and had no lid, so it was open. The milk was not homogenized, so the cream separated and rose to the top of the bottle. The birds learned that the cream was available and then to eat the cream from the bottle. Eventually, the milk producers began to put a foil cap on the bottles to keep the birds from eating the cream. However, the blue tits, but not the robins, learned to peck a hole in the foil caps and still get to the cream (see Figure 1.3).
[[Insert Figure 1.3, photo of bird pecking milk bottle cap, about here]]
_________________________________________________________
Think Ahead
*****Does the [birds’ milk-bottle pecking (or other behavior if a different example is used here)] behavior fit within the definition of learning? Before reading further, see if you can find examples of all the critical elements that we isolated in explaining the definition. Write down your answers.
________________________________________________________
To evaluate how well the example fits within the definition of learning, we can parse out the critical elements of the definition of learning and look for ways in which these elements apply to the behavior.
Inferred change: Although we cannot see the actual learning within the birds, we can infer that they learned to the change in behavior – particularly the blue tits, which showed two changes in behavior: originally learning to eat the cream from the bottles and subsequently learning to pierce the foil cap so that they could eat the cream. The blue tits provide us with a good example of the distinction between learning and performance: We cannot tell whether a bird has learned the behavior until it performs the behavior.
Organism’s mental state: As with most observed behaviors, we do not know what change has taken place within the bird’s brain, but it certainly appears that some internal change must have taken place.
Change that results from experience: Because the example of the birds learning to obtain cream from the milk bottles was not a controlled laboratory experiment, we are not certain of the exact learning mechanism involved in this process. However, it seems likely that this behavior is an example of observational learning in which an organism learns a behavior by watching and imitating the behavior of another organism – much as you might have learned how to eat using a fork or brush your teeth by watching your siblings or your parents.
Relatively permanent fashion: This behavior of stealing cream occurred over a period of many years. It has decreased today because of several factors: more people buying milk at supermarkets, people choosing to drink low-fat milk, and homogenization. An interesting question, of course, is whether the birds would quickly return to eating cream from milk bottles if all of these factors reverted back to the old ways of milk delivery. This principle implies that the birds should very quickly relearn how to get to the cream.
Potential for subsequent adaptive behavior: Getting to the cream was an adaptive behavior for the birds – it gave them a new and easy source of nutrition. Once they had adapted and learned to get the cream from the bottles, they had to adapt subsequently when the bottler began putting foil caps on the bottles. The blue tits managed to adapt to the foil caps whereas the red robins did not. de Geus (1997) attributed this difference to blue tits being a social bird species but red robins being more antagonistic toward each other. Thus, it appears that biological factors may be able to overwhelm the potential for learning.
Check Your Learning: Defining Learning
Learning is an example of adaptive behavior that promotes survival or is useful in some way.
Learning is an inferred change in the organism’s mental state that results from experience and that influences in a relatively permanent fashion the organism’s potential for subsequent adaptive behavior.
Learning is a hypothetical construct – a variable that psychologists believe in, use to explain events, and that cannot be directly observed.
The definition of learning makes the distinction between learning and performance an important one – scientists cannot observe learning directly, so they must gauge performance instead.
Many changes that can affect behavior are not considered learning, such as temporary changes, maturational development, and native responses.
Behaviors can be compared to the definition of learning to determine whether they should be considered examples of learning.
BIOLOGICAL CONSTRAINTS ON LEARNING
“Biological constraints on learning” is the formal name for the possibility raised in the last sentence before the Check Your Learning section: Biological factors may be able to overwhelm the potential for learning. Early learning psychologists believed in a principle rooted in behaviorism, the approach to psychology that views behavior as being influenced by an organism’s experience with the consequences of that behavior. This principle, equipotentiality, refers to the belief that learning does not vary as a function of the species, behavior, or conditions studied. In other words, behaviorists believed they could develop general laws of learning that would apply to all species, for all behaviors, under all conditions. However, over time, the assumption of equipotentiality has fallen on disfavor for a variety of reasons. We will lay out some general reasons for this change in this chapter; subsequent chapters will cover specific examples of biological constraints.
Seligman (1970) wrote an important paper that directly opposed equipotentiality. Instead, Seligman argued that learned behaviors fell into one of three categories depending on how biology interacted with learning. The three categories of learned behaviors Seligman proposed were prepared, unprepared, and contraprepared.
Prepared Behaviors
Prepared behaviors are those behaviors that organisms learn so easily and quickly that they almost appear to be instinctive. Often, they are vital to the survival of the organism. Learning theorists have long debated whether one-trial learning is possible; examples of behaviors learned in one or very few trials are likely prepared behaviors. Examples of prepared behaviors would be birds learning to fly or to sing. (You will learn about taste aversions – another example of a prepared behavior – in Chapter 5.) Both flying and singing are behaviors that birds “pick up” very easily. Although some people believe that these behaviors are instinctive, there are clearly elements of learning in each. As we mentioned earlier, birds fly very quickly once they have physically matured, but they do not immediately fly as well as mature birds. However, with some experience, their flying improves – a clear example of learning taking place.
Likewise, birds begin to sing with little apparent effort or learning – as long as they have grown up hearing other birds of their species sing. The example of bird song brings up another important biological constraint on learning: the notion of a critical period, a time in an organism’s development when a certain skill or ability is most readily acquired. Not only do birds have to hear other birds sing to develop normal song themselves, but they have to hear other birds sing during a specific time period when they are young (Nottebohm, 1970). If birds do not hear bird song during this period of time, then they will never develop a normal pattern of singing. It is critical that exposure to song occurs during this specific time – thus, the notion of a critical period. Bird song, therefore, appears to be a prepared behavior if learned during the critical period, but a contraprepared behavior if learning is attempted outside of the critical period.
Another example of a critical period comes from work on another phenomenon studied mainly in bird species, imprinting. Birds imprint, or appear to form an attachment, to moving stimuli after hatching. In the normal environment, the moving stimulus during that period would typically be the mother bird. Thus, baby birds would imprint on the mother and form the usual social attachment to other birds. However, researchers found that imprinting occurred only during a short time period shortly after hatching. For example, Hess (1959) found that imprinting in ducklings could occur only between 9 and 20 hours of age, with a peak at 13 to 16 hours (see Figure 1.4). Interestingly, researchers found evidence that imprinting is strongly influenced by biology, as birds have imprinted on a variety of objects that moved in their environment during the critical period. Konrad Lorenz is famous for a series of photographs taken after he substituted for a mother goose shortly after a batch of goslings hatched (see Figure 1.5). As the geese grew older, they preferred humans to other geese, even attempting to mate with humans.
[[Insert Figure 1.4, graph of imprinting success, about here]]
[[Insert Figure 1.5, photo of goslings following Konrad Lorenz, about here]]
Somewhat ironically, imprinting has been used with endangered species in the United Kingdom so that the nearly extinct Eurasian Cranes do not become attracted to humans (see Figure 1.6). Workers at the Wildfowl & Wetlands Trust (WWT) wear crane “costumes” and use litter pickers painted to look like adult cranes so that the baby cranes would imprint on cranes rather than humans. Using this process, workers hope that the cranes will mate with other cranes when they mature and thus reverse the extinction process. If successful, this procedure will allow the WWT to both provide human care for the cranes to help them survive and the biological imprinting process to help them thrive and mate.
[[Insert Figure 1.6, photo of litter picker painted to resemble a crane, about here]]
[[Insert Figure 1.7, Far Side cartoon of imprinting gone awry, about here]]
The biological concepts of critical periods and imprinting both have the capacity to make the learning process very simple. This outcome would result in behaviors that organisms are prepared to learn, in Seligman’s formulation. The third manner by which biology can influence prepared learning is a very simple biological concept – differences among species. For example, most of the imprinting work has focused on various bird species. In this work, the critical periods have been very small windows of time. Does imprinting occur in humans? How about critical periods?
________________________________________________________________
Think Ahead
*****Do you know of any examples of imprinting or critical periods in humans? Before reading further, write down your thoughts.
_________________________________________________________________
This question is a difficult one unless you have previously taken a course in developmental psychology. Some developmental psychologists (e.g., Bowlby, 1969) believe that human infants exhibit a behavior that develops in a fashion similar to imprinting. Known as attachment behavior, this behavior involves the infant developing an emotional bond with the caregiver. The primary reason behind this idea is that forming an infant-caregiver bond is critical to the survival of the infant. Thus, the infant will bond with the caregiver, usually the mother, after birth. Certainly the time for this process to occur would be much longer than that of avian species and is probably not as highly biological as imprinting in birds is – other factors also play a large role in the bonding process between infants and caregivers.
As for critical periods in humans, it seems that there may be a critical period for language acquisition in humans. If a child does not learn language by puberty (approximately), then the child will have an exceptionally difficult time ever learning language or becoming truly fluent in language. We will cover this topic in more detail at the end of this chapter.
Unprepared Behaviors
Unprepared behaviors are those that are learned in the process of what people think of as “normal or typical learning.” In unprepared behaviors, learning is somewhat difficult and takes some time to occur – it usually requires repeated trials, practice, and reinforcement. Seligman believed that most of the behaviors that psychologists study in the lab, such as classical conditioning and operant conditioning, fall into the unprepared category. In fact, most of the behaviors, concepts, and principles we will cover in this book fall into the unprepared category. These behaviors are typically not vital for survival, but they allow for adaptation to the environment or circumstances and thus are beneficial for the organism to learn. Lest you think we are ignoring two-thirds of all learned behaviors and learning situations in the book, unprepared behaviors account for the vast majority of learning situations, which is why we used the descriptions of normal and typical earlier in this paragraph. Because the rest of the book will cover the learning of unprepared behaviors, there is no reason to cover this category in any more detail at this point.
Contraprepared Behaviors
If you look up “contra” in your dictionary, you will find that it means “in opposition or contrast to” as a preposition and “on or to the contrary” as an adverb. In essence, contraprepared means anti-prepared or the opposite of prepared. Contraprepared behaviors, then, are those behaviors that are impossible or nearly impossible to learn despite extended training. Just as biology has made some behaviors (prepared) exceptionally easy to learn, biology has made contraprepared behaviors virtually unlearnable, typically because they conflict with another behavior (or behaviors) that is prepared and, often, vital to survival.
For example, it is easy to train pigeons to peck for food and to fly away when faced with danger. However, it is extremely difficult or impossible to train the opposite associations – to peck in the face of danger or to fly away to obtain food. As you can easily see, pigeons are biologically set to peck for food and fly away from predators, but not the opposite. Thus, pecking to avoid danger or flying to get food are likely contraprepared behaviors.
Keller and Marian Breland were early graduate students of B. F. Skinner and had traditional behavioral research training in graduate school, but they applied their skills to training animals to perform various behaviors for entertainment purposes such as trained animal acts, television, and movies. They founded Animal Behavior Enterprises in Hot Springs, Arkansas in the early 1940s for training animals and I.Q. Zoo in 1955 to showcase animals they had trained. Based on their experiences, Breland and Breland (1961) wrote an important paper about the difficulty of attempting to train animals to engage in contraprepared behaviors. Their graduate training included the equipotentiality principle we discussed earlier, so they fully expected that they could teach any animal to perform any trick using the principles of learning and reinforcement. However, they were in for a rude awakening.
As an example, the Brelands attempted to train a raccoon to pick up tokens and put them in a box in order to obtain a food reinforcement. The raccoon learned the behavior with some difficulty (as would be expected for an unprepared behavior) when it had to pick up only one token. However, given two tokens, the raccoon rubbed the tokens together repeatedly, even putting them partway into a slot on the box before pulling them out and rubbing them some more. The raccoon would rub the tokens together for minutes and became even more fixated on rubbing the tokens with repeated practice despite never receiving the food reinforcement.
_______________________________________________________
Think Ahead
*****Can you figure out what went wrong with the raccoon’s behavior with two tokens? Before reading further, writen down your thoughts.
______________________________________________________
Raccoons have a well-known reputation for “washing” their food before eating it. In reality, washing their food does not appear to be the true purpose for the behavior – raccoons will eat without washing or dipping their food in water. However, they do tend to rub or feel their food because their paws are quite sensitive, so they probably glean information from the rubbing behavior. The Brelands’ raccoon seems to have reverted to natural or instinctive behavior when it had two tokens. Why would the raccoon rub tokens, which are not edible, together? The raccoon probably associated the tokens with food because it used the tokens to procure food. Thus, the raccoon seemed to be contraprepared to use two tokens to gain food. With only one token, however, the raccoon had nothing to rub together, so natural behavior could not take over.
The Brelands also had difficulty teaching pigs to pick up a wooden token to deposit in a piggy bank (isn’t irony a funny thing?). At first, a pig would learn the behavior rather well, as it got a food reinforcement after dropping the “coin” in the bank. However, over time, the behavior of pig after pig deteriorated. The pigs would readily pick up the token, but then would drop it repeatedly on the way to the bank and push it with their snouts. The Brelands reported that eventually it would take a pig as long as 10 minutes to pick up four tokens and carry them a few feet to the bank.
_____________________________________________________________
*****Can you figure out what went wrong with the pigs’ behavior? Before reading further, write down your ideas.
_____________________________________________________________
You may know that pigs engage in rooting to find food – they dig with their snouts. The “misbehavior” of the pigs strongly resembled their natural food-gathering behavior. Because the tokens were associated with food, the pigs resorted to rooting them.
Given Seligman’s (1970) formulation regarding prepared, unprepared, and contraprepared behaviors and the evidence that this trichotomy does influence the learning process, it is clear that biological constraints can indeed influence the learning process. As we cover general principles of learning in the remainder of the book, it is important to remember that biology may influence those principles.
[[Insert Box 1.1, The Significance of Animal Research, about here]]
Check Your Learning: Biological Constraints on Learning
Behaviorists believed in the principle of equipotentiality – the idea that any organism can learn any behavior under any circumstance.
Seligman argued instead that biology negated the notion of equipotentiality and that behaviors fell into three categories with regard to learning based on biological preparedness. Prepared behaviors are learned very easily and quickly, almost instinctively; unprepared behaviors are learned with moderate difficulty and are typical of behaviors studied in the laboratory; contraprepared behaviors are very difficult or impossible to learn.
Another example of biology affecting comes in the notion of critical periods – time intervals during which an organism can learn a behavior; if not learned during this period, the behavior may not be learned.
Imprinting, a process by which an organism typically learns to identify with its species, seems to occur during a critical period.
Attachment behavior and language learning have the possibility of behaviors that humans learn during critical periods.
Animal research has been a cornerstone of learning studies. Behaviorists believed that they would develop general laws of learning, so the organisms with which they worked were irrelevant.
Learning in the Real World: Language Acquisition
It is always important to make sure that you do not lose sight of the forest because of the trees. Sometimes students in Learning courses get bogged down in learning theories and animal research studies and forget that the principles they are learning also apply to their lives. In an effort to make sure that you don’t lose the forest in the trees, we end each chapter with a real-life application section. These sections should help you see a principle from the chapter applied to the real world, perhaps even to your life.
Our real-world topic for this chapter is language acquisition: How do we learn to use language to convey and interpret meaning? As you might guess, language acquisition is not a simple process, so the answer will not be particularly simple either.
In terms of a “traditional” learning-oriented approach to language acquisition, behaviorists have viewed language just like any other learned behavior. B. F. Skinner (1957), in particular, accounted for language acquisition with the very same principles that he used to explain, for example, how a rat would learn to press a bar. We mentioned reinforcement earlier in the chapter – Skinner believed that parents and caregivers use reinforcement to help children learn to use language. Research from developmental psychology shows that babies go through a verbal stage known as babbling (Bukatko & Daehler, 2001), in which they produce all the sounds necessary to put together to make words. Behaviorists would explain that people hearing a baby babble would selectively reinforce sounds they heard – sounds that resemble words would draw excited responses from the listener, whereas sounds that do not resemble words would be ignored. Given that a multitude of research shows that reinforced responses tend to increase over time and nonreinforced responses tend to decrease over time, the baby would tend to make more and more utterances that sound like words and fewer that sound like gobbledygook. Factor in, also, the reinforcement gained when a baby wants to communicate that she wants a cookie and actually receives a cookie (rather than a diaper change), and you can see how reinforcement would be a powerful influence on communication.
We are not trying to imply that all of language acquisition is due to reinforcement and other straightforward learning principles. If you have taken a developmental psychology class, you may remember that there are also other theoretical perspectives on learning language: linguistic, nativistic, cognitive, social interaction, to name several. We are also not prepared to try to settle that issue here. However, we do want to return to a point that we mentioned briefly earlier in the chapter. There is some interesting information that points toward a biological constraint on language acquisition – the possible existence of a critical period for this important behavior. Lenneberg (1967) noted that learning a first or subsequent language is much easier during this hypothesized critical period, which seems to end around puberty for humans. In support of Lenneberg’s observation, older children learning a second language do not master some of the finer distinctions of the second language as well as younger children do (Lieberman, 1984). For example, it is difficult to speak the second language without an accent if it is learned after age 12 (approximately). These results about having difficulty with finer distinctions hold true for American Sign Language as well as spoken languages (Newport, 1990).
Although less clear-cut than studies previously mentioned, real-life case studies also seem to support the hypothesis of a critical period for language. “Genie” was a child who was locked in a room by her father, who was mentally ill, from about the age of 18 months until she was discovered at age 13 (Curtiss, 1977; Rymer, 1993). During this time, Genie had only minimal contact with people. When found, she was nearly mute and spoke only a couple of words. After many years of training by researchers at UCLA, she was able to communicate only at a very low level and never managed to master rules of syntax. There are other famous case studies that have shown similar findings (Bukatko & Daehler, 2001; Hetherington, Parke, Gauvain, & Locke, 2006). In all of these cases, there are competing hypotheses for the poor language skills, such as possible retardation or mental deficiency. However, they provide at least potential support for the idea of a critical period for language acquisition in humans.
Although it is not clear exactly how much of a role reinforcement and critical periods play in human language acquisition, it is clear that they have some influence. As we will see in the following chapter, this question does not lend itself to true experimental research, so it will not be a simple task to provide a definitive answer.
Box 1.1: The Significance of Animal Research
One thing you will notice as you go through this book is that a great deal of learning research has focused on animal learning; much of the research has used rats, dogs, pigeons, and other animals for subjects. If you are typical of many students in learning courses, you will wonder why there is so much emphasis on animal learning rather than on humans – you are likely much more interested in human learning than animals! To understand how this situation came about, we need to cover some historical information about behaviorism.
As you might remember from your introductory psychology course, John B. Watson is acknowledged as the father of behaviorism. Watson believed that in order to make psychology more scientific, the discipline should completely abandon studying the mind and consciousness (as in Freud’s psychoanalytic viewpoint). Instead, Watson argued that psychology should study only outwardly observable and measureable phenomena – in other words, behavior. It was Watson’s focus on behavior that gave this new approach its name of behaviorism. Watson was a strong believer in the power of the environment, rather than heredity, to influence behavior. One of his most famous quotes illustrates what he believed about the effects of environmental control:
Give me a dozen healthy infants, well-formed, and my own specified world to bring them up in and I’ll guarantee to take any one at random and train him to become any type of specialist I might select – doctor, lawyer, artist, merchant-chief, and yes, even beggar-man and thief, regardless of his talents, penchants, tendencies, abilities, vocations and race of his ancestors. I am going beyond my facts and I admit it, but so have the advocates of the contrary and they have been doing it for many thousands of years. Please note that when this experiment is made I am to be allowed to specify the way the children are to be brought up and the types of world they have to live in. (Watson, 1925, p. 82)
As you can probably tell from this quote, Watson subscribed to the tenet of equipotentiality, and it became an important concept of behaviorism. It is also important to know that behaviorists dominated the learning field for most of the 20th century. Thus, most researchers in the field believed in equipotentiality. A logical extension of the equipotentiality principle is that researchers believed that they would be able to uncover general laws of learning. If you believe that all organisms learn via the same principles, it makes sense to assume that you can develop laws of learning that will apply to all organisms. This position would also make sense given that behaviorists wanted to pattern psychology after other sciences – certainly the natural sciences have developed laws concerning their subject matter. Finally, because the behaviorists believed that all organisms learned using the same principles and that they would be able to develop laws of learning, they also believed that the species that they chose to study was not an important issue.
With these background assumptions, most learning researchers chose to pursue animal research – not because they were uninterested in human learning, but because they believed that human learning and animal learning were governed by the same principles. And for a number of reasons, it was simpler for these researchers to maintain animal laboratories than to work with human participants:
they would have easy access to their subject population whenever they wanted (researchers conducting human research must rely on their participants to remember to come to the lab),
they could more easily control the environment of their subjects (e.g., removing food so that food could serve as a reinforcement), and
they could study generations of subjects more quickly (e.g., rats reach sexual maturity in 5 weeks and have a gestation period of only 21 days, so it is a simple matter to study multiple generations in a short amount of time).
Although much of the research in this book will focus on animal learning, we urge you not to “tune out” on those topics – we will do our best to point out human analogies for the animal research findings. We also challenge you to look for human examples of findings that come from animal research. As you engage in this process, you will be learning the principles and concepts more fully and deeply – activities that will help you better understand and retain the material. Also, in our final chapters, we will pull together many of the principles from throughout the book and show you how they apply to your life and your learning.
Online Resources
Explore the case of the peppered moth in a lab exercise from a University of Michigan Global Change course syllabus.
(http://www.globalchange.umich.edu/globalchange1/current/labs/Lab7/Moth.htm)
Visit the blog of the Wildfowl & Wetlands Trust, the organization that works to save cranes and other migratory birds.
http://www.wwt.org.uk/blog/630/.html
Read further details about the lives and work of Keller and Marian Breland.
http://www.apa.org/monitor/2008/09/animals.html
http://www.house-of-learning.se/marianorbituary.htm
Read a 2008 news account of Genie’s story.
http://abcnews.go.com/Health/story?id=4804490
Key Terms and Definitions
adaptation
the ability of a species or organism to change over time to meet demands for survival
evolutionary adaptation
the change of biological traits over time for survival purposes; also known as natural selection
adaptive behavior
behavior that changes to promote survival or because it is useful to the organism; includes learning
learning
an inferred change in the organism’s mental state which results from experience and which influences in a relatively permanent fashion the organism’s potential for subsequent adaptive behavior
hypothetical construct
a variable that psychologists believe exists and that they use to explain events but cannot directly observe
long-term potentiation
a process by which a neural pathway is formed by increased neural excitability in synapses along the pathway
reinforcement
a consequence that follows a behavior that strengthens that behavior and increases its future probability
latent learning
learning that has taken place but not yet been exhibited
motivation
an internal process that energizes behavior, guides the organism’s behavior toward a goal, and maintains the behavior until the desired goal is achieved
emotion
physiological changes and conscious feelings of pleasantness or unpleasantness, aroused by external or internal stimuli, that lead to behavioral reactions
incentive
a property of reinforcement that increases or decreases its reinforcing property
maturational response
a behavior that becomes possible because an organism has grown older (developed)
native (innate) response
a behavior that an organism makes naturally without learning
behaviorism
the approach to psychology that views behavior as being influenced by an organism’s experience with the consequences of that behavior
equipotentiality
belief that learning does not vary as a function of the species, behavior, or conditions studied
prepared behaviors
those behaviors that organisms learn so easily and quickly that they almost appear to be instinctive
unprepared behaviors
those behaviors that are learned in the process of what people think of as “normal or typical learning”
contraprepared behaviors
those behaviors that are impossible or nearly impossible to learn despite extended training
critical period
a time in an organism’s development when a certain skill or ability is most readily acquired
imprinting
phenomenon in which an organism appears to form an attachment; e.g., birds imprint to moving stimuli after hatching
attachment behavior
phenomenon in which an infant develops an emotional bond with the caregiver
References
Biological Sciences Curriculum Studies. (2002). Biology (9th ed.). Dubuque, IA: Kendall Hunt.
Bowlby, J. (1969). Attachment and loss (Vol. 1: Attachment). New York: Basic Books.
Breland, K., & Breland, M. (1961). The misbehavior of organisms. American Psychologist, 16, 681-684.
Bukatko, D., & Daehler, M. W. (2001). Child development: A thematic approach (4th ed.). Boston: Houghton Mifflin.
Cook, J. L., & Cook, G. (2005). Child development: Principles and perspectives. Boston: Pearson.
Curtiss, S. (1977). Genie: A psycholinguistic study of a modern-day “wild child.” New York: Academic Press.
de Geus, A. (1997). The living company. Boston: Harvard Business School Press.
Hess, E. H. (1959). Imprinting. Science, 130, 133-141.
Hetherington, E. M., Parke, R. D., Gauvain, M., & Locke, V. O. (2006). Child psychology: A contemporary viewpoint (6th ed.). Boston: McGraw-Hill.
Hilgard, E. R., & Bower, G. H. (1975). Theories of learning (4th ed.). Englewood Cliffs, NJ: Prentice-Hall.
Hillner, K. P. (1978). Psychology of learning: A conceptual analysis. New York: Pergamon Press.
Kimble, G. A. (1967). The definition of learning and some useful distinctions. In G. A. Kimble (Ed.), Foundations of conditioning and learning (pp. 82-99). New York: Appleton-Century-Crofts.
Lenneberg, E. H. (1967). Biological foundations of language. New York: Wiley.
Lieberman, P. (1984). The biology and evolution of language. Cambridge, MA: Harvard University Press.
Navarick, D. J. (1979). Principles of learning: From laboratory to field. Reading, MA: Addison-Wesley.
Newport, E. L. (1990). Maturational constraints on language learning. Cognitive Science, 14, 11-28.
Nottebohm, F. (1970). The ontogeny of birdsong. Science, 167, 950-956.
Rymer, R. (1993). Genie: A scientific tragedy. New York: HarperCollins.
Seligman, M. E. P. (1970). On the generality of the laws of learning. Psychological Review, 77, 406-418.
Skinner, B. F. (1957). Verbal behavior. New York: Appleton-Century-Crofts.
Tarpy, R. M., & Mayer, R. E. (1978) Foundations of learning and memory. Glenview, IL: Scott, Foresman and Company;
Tolman, E. C., & Honzik, C. H. (1930). Introduction and removal of reward, and maze performance in rats. University of California Publications in Psychology, 4, 257-275.
Watson, J. B. (1925). Behaviorism. New York: W. W. Norton.
Table 1.1
Hillner’s (1978) List of Diverse Phenomena Associated With Learning
Learning encompasses both animal and human behavior. It is applicable to the behavior of intact or whole organisms, and even to the adaptive behavior of inanimate model systems such as computer simulations.
Learning involves events as diverse as the acquisition of an isolated muscle twitch, a prejudice, a symbolic concept, or a neurotic symptom.
Learning is not limited to the external responses of the organism, but also to internal physiological concepts.
Learning is concerned with the original acquisition of a response or knowledge, with its later disappearance (extinction), its retention over time (memory), and its possible value in the acquisition of new responses (transfer of training).
Learning is related to such nonlearning phenomena as motivation, perception, development, personality, and social and cultural factors.
Learning has a physical reality (physiological, biochemical) as well as a strictly psychological (functional) reality.
Learning deals with the behavior of the average subject and with individual differences among subjects.
The study of learning is associated with a long academic and scholarly tradition, but also serves as a source of practical application and technology.
The learning process is continuous with, and a component of, the more general linguistic, cognitive, information-processing, and decision-making activities of the organism.
Note. From Hillner (1978), pp. 1-2
Figure 1.1
Peppered Moths
http://www.truthinscience.org.uk/site/content/view/127/65/
Figure 1.2
Results from Tolman and Honzik’s (1930) Experiment
Adapted from:
Tolman, E. C., & Honzik, C. H. (1930). Introduction and rremoval of reward, and maze performance in rats. University of California Publications in Psychology, 4, 257-275.
Figure 1.3
Birds learned how to get cream from the top of milk bottles despite the bottles having a cap
http://www.bbc.co.uk/blogs/monthwithoutplastic/blue_tit_colinsargent.jpg
Figure 1.4
Figure 1.5
Goslings imprinted on Lorenz
http://2.bp.blogspot.com/_2i0N8q9acw8/R3aJj9Be18I/AAAAAAAABvA/QC82SLAIds8/s400/LorenzAndGeese.jpg
Figure 1.6
Using the principles of imprinting to prevent endangered species from imprinting on humans
http://www.newscientist.com/blog/environment/2007_06_01_archive.html
http://www.wwt.org.uk/blog/630/.html
Figure 1.7 (?)
