CHAPTER 6 LEARNING OUTLINE I LEARNING ABOUT STIMULI

CHAPTER 11 OECD AVERAGE AND OECD TOTAL BOX
 CONTENTS PREFACE IX INTRODUCTION 1 REFERENCES 5 CHAPTER
 NRC INSPECTION MANUAL NMSSDWM MANUAL CHAPTER 2401 NEAR‑SURFACE

32 STAKEHOLDER ANALYSIS IN THIS CHAPTER A STAKEHOLDER ANALYSIS
CHAPTER 13 MULTILEVEL ANALYSES BOX 132 STANDARDISATION OF
CHAPTER 6 COMPUTATION OF STANDARD ERRORS BOX 61

CHAPTER 6


CHAPTER 6

Learning

outline

I. LEARNING ABOUT STIMULI

Habituation is considered a simple form of adaptive learning; organisms stop paying attention to stimuli that are often repeated and that do not signal any important environmental events.

According to the opponent-process theory, habituation to repeated stimuli causes two processes. The first, known as the A-process, is an almost reflexive increase or decrease in some response. The second, called the B-process, causes an opposite or opposing response. The opponent-process theory has been used to explain drug-tolerance development.

II. CLASSICAL CONDITIONING: LEARNING SIGNALS AND ASSOCIATIONS

A. Pavlov’s Discovery

1. Ivan Pavlov’s experiment was the first demonstration of classical conditioning. Pavlov’s experiment had three phases. During the first phase, a natural reflex and a neutral stimulus were established. During the second phase, the neutral stimulus and the stimulus causing the natural reflex were repeatedly paired. During the third phase, the neutral stimulus alone caused some form of the natural reflex to occur.

2. In classical conditioning, a neutral stimulus is paired with a stimulus that elicits a reflex or other response until the formerly neutral stimulus alone elicits a similar response. The stimulus that causes the natural reflex is called the unconditioned stimulus (UCS); the reflex itself is designated the unconditioned response (UCR). The neutral stimulus that is paired with the UCS is the conditioned stimulus (CS), and the learned response to the conditioned stimulus is the conditioned response (CR).

B. Conditioned Responses over Time: Extinction and Spontaneous Recovery

The CS will continue to elicit a CR only if the UCS continues to appear at least some of the time. If the CS and UCS are unpaired (that is, the CS is not followed by the UCS), the CR gets weaker and weaker and undergoes extinction. However, the learned relationship between the CS and the UCS is not completely forgotten, as is demonstrated by reconditioning and spontaneous recovery. In reconditioning, the relationship between the CS and the UCS is relearned as the stimuli are paired once again. This time the CS will elicit the CR much more quickly. Spontaneous recovery occurs when, after no presentation of either the CS or the UCS for a period of time, a single presentation of the CS elicits the CR.

C. Stimulus Generalization and Discrimination

Stimulus generalization occurs when a stimulus similar but not identical to the original stimulus also elicits a response. Stimulus discrimination is a complementary process through which organisms learn to differentiate between stimuli that are similar but not identical to the CS.

D. The Signaling of Significant Events

In classical conditioning, the CS acts as a signal that the UCS is about to appear. Situations that highlight and strengthen the CS-UCS relationship will produce stronger CRs.

1. Timing. Classical conditioning produces the strongest CRs when the CS precedes the UCS by no more than a few seconds.

2. Predictability. A strong CR will be developed if a very noticeable CS is reliably followed by the UCS.

3. Signal Strength. The relationship between the CS and the UCS is learned faster as the salience or intensity of the CS and UCS increases.

4. Attention. Often, more than one CS is associated with a UCS. The CS that is most attended to will be the best predictor of the UCS.

5. Second-Order Conditioning. Second-order conditioning occurs when a second conditioned stimulus predicts the presence of the first conditioned stimulus, which predicts the presence of the UCS.

6. Biopreparedness. Humans and animals may be innately likely or biologically “prepared” to learn certain adaptive associations. Nausea is likely to be a conditioned response to an internal stimulus such as taste (conditioned taste aversion), and pain is likely to be a conditioned response to an external stimulus such as noise.

E. Some Applications of Classical Conditioning

1. Phobias. Phobias are fears of objects or situations that are not harmful. Classical conditioning can both produce (via stimulus generalization) and eliminate (through systematic desensitization) phobias.

2. Predator Control. Some ranchers have set out mutton laced with lithium for wolves and coyotes. The dizziness and severe nausea (UCR) caused by the lithium becomes associated with the smell and taste of mutton (CS), thus making sheep an undesirable meal.

3. Detecting Explosives. Insects can detect explosive material. By repeatedly pairing the taste of sugar water with the smell of chemicals used in some explosives, wasps develop a conditioned response to the smell alone, displaying an attraction to it.

4. Predicting Alzheimer’s Disease. A puff of air directed at your eye is a UCS that causes you to blink. A light preceding the air puff can elicit a CR of a blink. This CR is impaired in individuals likely to develop Alzheimer’s, allowing for early identification and medication.

III. OPERANT CONDITIONING: LEARNING THE CONSEQUENCES OF BEHAVIOR

People learn more than just an association between neutral and unconditioned stimuli. For many behaviors, the stimuli that follow an action are important. In other words, people learn to respond in a way that brings about positive consequences.

A. From the Puzzle Box to the Skinner Box

According to the law of effect, if a response made in the presence of a particular stimulus is followed by a reward, that response is more likely to be made the next time the stimulus is encountered. Responses that are “instrumental,” meaning that they help produce some rewarding or desired effect, are learned; therefore, Thorndike called this learning instrumental conditioning. Skinner’s emphasis on how an organism learns to “operate on” its environment to produce a positive effect led him to rename instrumental conditioning to operant conditioning.

B. Basic Components of Operant Conditioning

1. Operants and Reinforcers. An operant is a behavioral response that has some effect on an organism’s environment. In operant conditioning, people learn the relationship between operants and their consequences. A reinforcer is a consequence that increases the probability that a behavioral response will occur again. Two types of reinforcers exist. Positive reinforcers are positive stimuli that act like rewards. Negative reinforcers are negative stimuli that, once removed, encourage or reinforce behavior.

2. Escape and Avoidance Conditioning. Negative reinforcements are used in escape and avoidance conditioning. In escape conditioning, an organism learns behaviors that lead to an escape from an unpleasant situation (negative reinforcement). In avoidance conditioning, an organism learns behaviors that allow it to completely avoid an unpleasant situation (negative reinforcement). Avoidance conditioning is very strong and may prevent an organism from learning new behaviors.

3. Discriminative Stimuli and Stimulus Control. Discriminative stimuli signal to an organism that reinforcement is available if a certain response is made. This response is said to be under stimulus control. Stimulus generalization, recognizing a stimulus similar to the original stimulus that signaled reinforcement, also occurs in operant conditioning.

C. Forming and Strengthening Operant Behavior

1. Shaping. Creation of new responses never before displayed can be accomplished through shaping, or reinforcing successive approximations (more and more like the desired response) of the desired behavior.

2. Secondary Reinforcement. Often, operant conditioning will begin with primary reinforcers—events or stimuli that are intrinsically rewarding, such as food. A secondary reinforcer is a previously neutral stimulus that, if paired with a stimulus that is already reinforcing, will itself take on reinforcing properties.

3. Delay and Size of Reinforcement. Operant conditioning is strongest when the delay in receiving a reinforcer is short and when the reinforcer is large.

4. Schedules of Reinforcement. On a continuous reinforcement schedule, every correct response receives a reward. On an intermittent or partial reinforcement schedule, reinforcement is received only some of the time.

a) Fixed-ratio (FR) schedules give a reward after a fixed number of responses.

b) Variable-ratio (VR) schedules give a reward after an average number of responses.

c) Fixed-interval (FI) schedules reward the first response displayed after a fixed time interval.

d) Variable-interval (VI) schedules reward the first response displayed after a varying time interval.

5. Schedules and Extinction. Eliminating reinforcers for behavioral responses eventually causes the behavioral response to cease (extinction). The partial reinforcement extinction effect demonstrates that it is more difficult to extinguish an operant behavior learned under a partial rather than a continuous reinforcement schedule.

D. Why Reinforcers Work

Primary reinforcers are items that fulfill basic needs or are inherently rewarding experiences, such as relief from pain.

The Premack principle asserts that each person has a hierarchy of behavioral preferences and that the higher an activity is in that hierarchy, the greater its reinforcement power.

The disequilibrium hypothesis contends that any activity can become a reinforcer if access to it has been restricted for a time.

Biological psychologists have found that stimulation of “pleasure centers” in the brain is a powerful reinforcer, suggesting a physiological component to reinforcement.

E. Punishment

Punishment presents an aversive stimulus or removes a pleasant stimulus to decrease the frequency of a behavior. Punishment has several disadvantages. First, it does not eliminate learning; it merely suppresses a behavior. If an organism knows that punishment is unlikely, then the behavior is repeated. Second, punishment is not effective unless it immediately follows the undesired behavior. Third, punishment may be associated with the punisher so that eventually the punisher is feared. Fourth, the organism being punished may learn to relate to others in an aggressive manner. Fifth, punishment makes clear what behaviors are incorrect, but it doesn’t provide any demonstration of desired behaviors.

Punishment can work if used wisely. One should punish the behavior, not the person; punish immediately; use a severe enough punishment to eliminate the behavior; and explain and reinforce more appropriate behaviors.

F. Some Applications of Operant Conditioning

Operant conditioning can be used to teach people the “rules” of social behavior and to eliminate problematic behavior and reinforce positive, desired behavior in people afflicted with mental retardation, autism, and other behavior disorders. Operant conditioning principles can also be used to help people understand the stimuli that trigger behaviors they want to eliminate (such as smoking or overeating). Understanding and avoiding discriminative stimuli can reduce the undesired behavior.

G. Linkages: Neural Networks and Learning

As the chapters on perception and memory suggest, neural networks likely play an important role in storing and organizing information. Psychologists have developed models of how these associations that occur in learning might be established as neural connections. These parallel distributed processing models suggest that knowledge is distributed throughout a network of associations. They focus on how these connections develop through experience. The weaker the connection between two items, the greater the increase observed in connection strength when they are experienced together.

IV. COGNITIVE PROCESSES IN LEARNING

Cognitive processes, such as how people store, represent, and use information, can influence learning.

A. Learned Helplessness

Humans and animals placed in situations that eliminate their control over the environment tend to give up any effort to exert control over their environment in new situations. They learn to be helpless. Learned helplessness is a tendency to give up any effort to control the environment.

B. Focus on Research Methods: An Experiment on Human Helplessness

In this experiment the dependent variable—the degree to which subjects acted to control noise—could be affected by two independent variables: prior experience with noise and expectation about the ability to influence the noise. People tend to feel less able to control a situation if they previously could not control it or if they are told they are powerless. These experiences may give rise to negative thinking or pessimistic explanatory styles, which may affect performance and mental health. Repeated experience with successful control may develop an optimistic cognitive style.

C. Latent Learning and Cognitive Maps

Learning that is not immediately evident in an organism’s behavior is known as latent learning. Cognitive maps are mental representations of the environment.

D. Insight and Learning

The cognitive process of insight involves understanding the global organization of a problem.

E. Observational Learning: Learning by Imitation

A series of experiments by Albert Bandura demonstrated that people learn by watching others, which is termed observational learning or social learning. In observational learning, a person learns new behaviors by watching others’ behavior. In vicarious conditioning, a type of observational learning, a person learns new behaviors by observing the consequences (reinforcement or punishment) of someone else’s behavior.

F. Thinking Critically: Does Watching Violence on Television Make People More Violent?

What am I being asked to believe or accept?

Watching violence on television causes violent behavior in viewers.

What evidence is available to support the assertion?

Anecdotes and case studies describe incidents of violence among children after watching violence on TV. Many studies show a positive correlation between watching and doing violence. Controlled studies show increases in violent behavior following violent visual stimulation.

Are there alternative ways of interpreting the evidence?

Anecdotal evidence and correlational evidence do not prove causation. It may be possible that the dependent variables used in the controlled studies are not similar enough to violence in everyday life.

What additional evidence would help to evaluate the alternatives?

More controlled studies are needed to better understand the relationship between viewing violence and committing violent acts. However, ethical considerations prohibit these types of studies.

What conclusions are most reasonable?

The large number of studies done in this area make it reasonable to conclude that watching TV violence may be one cause of violent behavior.

V. USING RESEARCH ON LEARNING TO HELP PEOPLE LEARN

A. Classrooms Across Cultures

Studies show that the average academic performance of U.S. students tends to lag behind that of students in other countries. Suggested causes include static classroom formats and less time spent on practice.

B. Active Learning

Active learning exercises improve memory of material and make classrooms more enjoyable. Active learning is a more elaborate processing in which students solve problems in groups, think about how material relates to what they know, and answer every question asked by the teacher.

C. Skill Learning

Practice is the most critical component of skill learning. Practice should continue until the skill can be performed automatically for perceptual-motor skills. For cognitive skills, practice should focus on retrieval of information from memory.

Feedback lets the learner know if she or he is correct and may provide understanding of the cognitive and physical processes used in the skill. Feedback should be given after a learner is finished practicing and should not be so detailed that the learner does not have to learn from his or her own mistakes.



CONFIGURING USER STATE MANAGEMENT FEATURES 73 CHAPTER 7 IMPLEMENTING
INTERPOLATION 41 CHAPTER 5 INTERPOLATION THIS CHAPTER SUMMARIZES POLYNOMIAL
PREPARING FOR PRODUCTION DEPLOYMENT 219 CHAPTER 4 DESIGNING A


Tags: learning outline, skill learning, learning, stimuli, about, outline, chapter