Week 3: Learning

1
Learning Objectives:

What is learning?

How does classical conditioning explain learning?

How new behaviours are learned?

What is Social Learning?

1.1 What is learning?

Learning results from our experience and interaction with the objects and the
events of our environment. It is defined as ‘as a process through which experience
produces a lasting change in behavior or mental processes’.

Without the ability to form new knowledge our lives would rely entirely on
reflexes and on the innate (inborn) behaviors (instincts). More complex learning
behaviors are evolutionary advances over instincts, allowing prompt adjustments to
the environment (Zimbardo, Weber, & Johnson, 2003, pp. 206).

1.2 Simple forms of learning

 Habituation is the form of learning that leads to the elimination of a

response towards a stimulus after a repeated exposure (e.g. if you live near a
busy street, you may learn to ignore the sound of the traffic)

2
  Mere exposure effect is the tendency to prefer familiar stimuli (this is a
technique highly used in advertising).

1.3 Complex forms of learning

 Classical Conditioning
 Operant Conditioning

2.1 Classical Conditioning (CC)

Classical Conditioning is a form of learning where a specific stimulus that

produces an innate reflex is associated with a neutral stimulus, resulting in
the elicitation of the same response.

2.1.1. Fundamental features of CC

The Classical Conditioning approach was firstly described through the experiments
of Ivan Pavlov and his research team.

The experiment:

Pavlov and his team used an untrained dog. In intervals, a tone was sounded,
and the dogs were given food. The first reaction to the tone was an orienting
response (stretching of the ears and turning the head towards the source of the
sound). Repeated pairings between the tone and the food, the orienting
response stopped, and the dogs began to salivate in response to the tone alone.

Pavlov discovered that a neutral stimulus when paired with a reflex-producing

stimulus, elicits a learned response that is similar to the original reflex.

3
Main components of Classical Conditioning:
 Unconditioned stimulus (UCS): stimulus that activates a reflexive
response (e.g. food provokes salivation reflex)
 Unconditioned response (UCR): the response elicited by an
unconditioned stimulus without prior learning (e.g. salivation at the sight of
food). No learning is involved between UCS and UCR
 Acquisition: is the first stage of classical conditioning. A neutral stimulus
(e.g. a tone) is repeatedly paired with a UCS. After several trials, the
neutral stimulus elicits the same response as does the UCS
 Conditioned Stimulus (CS): a (previously) neutral stimulus which elicits a
conditioned response (e.g. the tone produces salivation).
 Conditioned Response (CR): a response elicited by a previously neutral
stimulus that has become associated with the unconditioned stimulus.

In Classical Conditioning, proper timing of the pairing between the two stimuli is
essential. In most cases, the CS and UCS must be presented contiguously (closely
together) so that the connection can be made (Zimbardo, Weber, & Johnson, 2003,
pp. 209).

Classical Conditioning

UCS (Food) UCR (Salivation)

NS (Bell) UCS (Food) UCR (Salivation)

CS (Bell) CR (Salivating)

4
Figure 1 Illustrates Pavlov’s Classical Conditioning experiment. Step one food was
presented to the dog (causing salivation). Food was presented, while the tuning
fork was struck. This act was paired and repeated several times. The tuning fork
was struck without the food presented and salivation occurred. Conditioning was
retrieved from https://psychpics.com/2015/12/11/behaviorism-i-historical-progress/

Extinction and Spontaneous Recovery

Every conditioned response does not permanently stay in the behavioral repertoire.
Extinction occurs when a conditioned response is repeatedly presented without the
simultaneous presentation of the UCS. In some cases a spontaneous recovery may
appear. This is the reappearance of the conditioned response after the extinction
(the extinction actually causes a suppression of a conditioned response and not a
full elimination).

5
Generalization
Generalization is a process where the conditioned response occurs to stimuli
similar with the CS (e.g. dogs salivate to sounds similar to the conditioned
stimulus).

Discrimination
Discrimination is the opposite of generalization. An organism learns to respond to
one stimulus but not to stimuli that are similar (e.g. the dogs distinguish between
different sounds). Practically, this was done by presenting one tone where food
followed, whereas another tone was not.
(Zimbardo, Weber, & Johnson, 2003, pp. 208-212)
2.1.2 Applying Classical Conditioning
Pavlov’s experiments referred to appetitive conditioning (pleasant stimulus e.g.
food). Classical conditioning also involves unpleasant or aversive stimuli such as
painful electric shocks.

Aversive conditioning occurs when the conditioned stimulus promotes an aversive

response. Typically in such cases, occurs a defensive reflex response (e.g. fainting
at the sight of blood). A generalized fear reaction involving the autonomic nervous
system is involved (e.g. heart rate).

2.1.3 The Case of Little Albert

Conditioned fear was first demonstrated by John Watson and Rosalie Rayner. The
researchers conditioned an infant named Albert to react fearfully to a white rat.
The child’s fear was created by repeatedly presenting the rat paired with a loud
sound. After seven trails, the child started reacting with fear at the appearance of

6
the rat alone. After the establishment of the learned fear, the researchers showed
that his fear was generalized to other furry objects (e.g. Santa Claus mask).
(Zimbardo, Weber, & Johnson, 2003, pp. 213)

3.1 Operant Conditioning – Shaping New Behaviors

According to Operant Conditioning, behavior change can be affected by the

consequences that follow the behavior (rewards and punishments). These ideas
were first described by Edward Thorndike and enhanced by B.F. Skinner.

Defining Operant Conditioning:

Rewards and punishments are consequences that influence the likelihood of

the behaviours they follow recurring in the future.

 Rewards: money, praise, food or high grades.

 Punishments: pain, loss of privileges, or low grades

Operant conditioning is a more powerful form of learning because:

1. It involves a wider spectrum of behaviors
2. It accounts for new and complex behaviors (not just innate reflexes like in
Classical Conditioning).

3.1.2 Trial and Error

Edward Thorndike studied cats and the way they could escape from ‘puzzle
boxes’. Thorndike put food outside the puzzle box, and the cats had to open the
door of the box in order to reach the food. To solve this problem, the cats used trial
and error learning. Based on the consequences that followed their responses, the

7
cats’ behavior was shaped. The researcher believed that his observations were
equally applied to human learning.

Trial and Error: The learner gradually discovers the correct response by
attempting many behaviours and noting which ones produce the desired
consequences

Figure 2 shows Thorndike’s Puzzle Box. Initially, a fish was put outside the
box. The cat pushed the bar, unlocking the gate, then pulled the string managed
to unlock the gate. Finally, the cat pushes the gate down
https://psychpics.com/2015/12/11/behaviorism-i-historical-progress/

3.2 Skinner and Operant Conditioning

Skinner was influenced by Thorndike’s views about how environmental
consequences can affect the occurrence of the behavior. He called his system

8
‘operant conditioning’ because it focuses on operant responses (operant is an
observable, voluntary behavior that an organism emits to ‘operate’ or have an
effect on the environment).
(Zimbardo, Weber, & Johnson, 2003, pp. 218)

3.2.1 Reinforcement
Reinforcer is a condition (either by presenting or removing a stimulus), that occurs
after a response and affects the appearance of the behavior.

 Positive Reinforcement: is a stimulus presented after a behavior and

increases the probability of that response occurring again (food, money,
praise smile etc.) (Positive means adding).

 Negative Reinforcement: the removal of an unpleasant or aversive

stimulus after a behavior (Negative means removing) (e.g. using an
umbrella to avoid getting wet).

3.2.2 The Skinner box

Skinner invented a box with a lever that an animal could press to reach food.

Skinner box: An operant chamber that can be programmed to deliver

reinforcers and punishers contingent on an animal’s behaviour

9
 Reinforcer
s

Primary Reinforcer (e.g. Secondary Reinforcer or

food and sex) have a Conditioned Reinforcers
biological value to an (e.g. tokens, money)
organism acquire reinforcing
power by a learned
association with a
primary reinforcer

3.2.3Continuous VS Intermittent Reinforcement

 Continuous Reinforcement: A type of reinforcement which occurs after all

of the correct responses. The failure to reward a correct response on one trial
can be misinterpreted as an incorrect response. This form of reinforcement
is useful for shaping complex new behaviors, such as playing a musical
instrument (it gives the learner a feedback that performance has improved).

 Intermittent Reinforcement (or partial reinforcement): A type of

reinforcement where some but not all, correct responses are being
reinforced. This is a less frequent schedule of reward (e.g. after every third
correct response) that can still serve as an incentive to maintain the behavior.
This is considered as a more effective way to maintain the behaviors.

10
3.2.4 Scheduling Reinforcement
Reinforcement has two main forms. The ratio schedule rewards a subject after a
certain number of responses, whereas the interval schedule provides a reward after
a certain time interval.

Ratio Schedules: Rewards are distributed based on the number of response (e.g.
employees get paid based on the amount of work they perform).
This is distinguished into two types: fixed ratio and variable ratio schedules
 Fixed ratio Schedules: Programs by which reinforcement is provided
after a certain number of responses (e.g. an employer who pays the
workers a dollar for every ten cartons of products they manufacture).
 Variable Ration Schedules: Reinforcement programs by which the
number of responses required for reinforcement varies from trial to trial
(e.g. the slot machine players).

Interval Schedules: A program by which reinforcement depends on the time

interval elapsed since the last reinforcement. This is distinguished into two kinds:
fixed intervals and variable intervals schedules.
 Fixed Intervals: Programs by which reinforcement is contingent upon a
certain fixed time period (e.g. monthly paycheck). This results in a low rate
of responding.
 Variable Interval Schedules: Programs by which the time period between
reinforcements varies from trial to trial. The resulting rate of responding can
be high (e.g. random visits by the boss delivering reinforcement in the form
of praise).
(Zimbardo, Weber, & Johnson, 2003, pp. 221)

11
Summary
Schedule of Reinforcement Advantages and Disadvantages

Behavior changes quickly. Extinction

Fixed Ratio
occurs quite rapidly when rewards cease

Behavior changes quickly. Extinction

Variable Ratio
occurs slowly when rewards cease

Behavior changes more slowly.

Fixed Interval
Extinction occurs quite rapidly when
rewards cease

Behavior changes more slowly.

Variable Interval
Extinction occurs quite slowly when
rewards cease

3.2.5 Does Punishment work?

Punishment: An aversive stimulus which occurs after a response and

diminishes the strength of that response

 Positive Punishment: The application of an aversive stimulus (e.g.

touching a hot plate causes pain) after a response and this leads to a
reduction of the possibility of repeating the same behavior.

12
  Negative Punishment: The removal of an attractive stimulus after a
response (e.g. taking the toy away from a misbehaving child).

In order to be effective, punishment must be administered consistently. Intermittent

punishment is far less effective than punishment delivered after every undesired
response. In fact, not punishing an occurrence of unwanted behavior can have the
effect of rewarding it (e.g. supervisor overlooking an employee from coming to
work late).
(Zimbardo, Weber, & Johnson, 2003, pp. 225)

Behavior and Consequences

Positive: Apply (add) a stimulus Negative: Remove (Subtract) a stimulus

Negative Reinforcement
Positive Reinforcement
Example: Your headache disappears after
Example: An employee gets a bonus for
taking an aspirin. You take an aspirin again
good work and is motivated to keep
the next time you have a headache
working hard
Negative Punishment
Positive Punishment
Example: A child stayed out late and
Example: A speeder gets a traffic ticket,
misses dinner. Next time the child comes
then drives away more slowly
home on time

13
2.5.6 Consequences of punishment
Societies rely on punishment and the threat of punishment to keep people in order
(e.g. putting people in jail, giving people a fine). Punishment often has an
immediate result and a change in behavior.
Then why do punished children continue to misbehave? Why do employees
continue to arrive late at work? Punishment does not always work the way it is
intended to.

1. The power of punishment disappears when the threat of punishment is

removed (e.g. drivers will observe the speed limit when they know the
highway patrol is watching). Punishment only works when there is the
constant control of the environment
2. Punishment often triggers aggression. Punished organisms usually seek to
flee from a punishment. But if all escape is blocked, they may become
aggressive
3. An organism that is constantly being punished eventually may give up any
attempt to flight or fight. The passive acceptance of punishment is a
behavior pattern called learned helplessness (this may lead to depression)
4. Punishment makes the punished person apprehensive. This inhibits learning
new and better responses and turns the attention only to the punished action.
5. Punishment is often applied unequally (e.g. boys are punished more often
than girls)
(Zimbardo, Weber, & Johnson, 2003, pp. 225)

14
 Effective Punishment:

 Should be swift and brief, a delay will decrease its effectiveness

 Should be certain, administered every time the unwanted response
occurs
 Should be limited in duration and intensity
 Should clearly target the behaviour, not the character of the person
 Should be limited to the situation in which the response occurred
 Should not give mixed messages to the punished person (e.g. a child is
punished because he hit another child but the parent is allowed to hit
the child)
 Is usually negative punishment, consisting of penalties (such as loss of
privileges) instead of the application of unpleasant stimuli (e.g. pain)

(Zimbardo, Weber, & Johnson, 2003, pp. 225)

Zimbardo,
3.1 The Cognitive 2003, pp. 225)
Revolution
Cognitive psychologists disagreed with the idea of learning being a simple
connection between a stimulus and a response as supported through Classical and
Operant conditioning. A new perspective of learning emerged through a more
cognitive point of view. On a neural level, learning involves a physical change that
strengthens the synapses in groups of nerve cells (a process called long-term
potentiation).
(Zimbardo, Weber, & Johnson, 2003, pp. 236)

15
3.1.2 Gestalt Psychology (Kohler):

The Gestalt theory supports that perception (the way a stimulus is being received
and interpreted), affects the responses. In line with this basic idea, the Gestalt
theorists argue that people use their insight in order to solve a problem or
determine their response to stimuli (in contrast with behaviorists). Wolfgang
Kohler observed chimpanzees and identified that some apes learned more quickly
than other, how to reach their food. Despite the fact that the chimpanzees used trial
and error strategies, these attempts were not merely enhanced by the reinforcement
of their prior attempt. The researcher believed that the animals used their prior
attempts to determine their next attempt. This means that the apes used insight to
solve the problem (Blosser, 1973 in http://hci.liacs.nl/files/gestalt-excerpt.pdf).

The Gestalt approach considers problem-solving approaches (and subsequently

learning) as the culmination of mental processes that create representations of the
problem situation.

3.1.3 Social Learning (Bandura):

Another learning approach is known as Social Learning. Based on the views of the
pioneer of this approach Albert Bandura, learning is a cognitive process that occurs
in the social environment and can be influenced through observation or direct
instruction. This can happen in the absence of motor reproduction or direct
reinforcement (Bandura & Walters, 1977). In addition, learning can also happen
through the observation of rewards and punishments, a process known as vicarious
reinforcement (Bandura, 1971).

16
In brief, in Bandura’s classic experiments, children watched videos with adults
punching, hitting and kicking an inflated plastic clown (BoBo Doll Experiment).
Children showed similar aggressive behavior towards the doll. Significantly, these
children were more aggressive than those in a control condition who had not
observed the aggressive models. According to Bandura’s study, imitative learning
happens in situations where individuals cannot rely on their personal experiences
(Zimbardo, Weber, & Johnson, 2003, pp. 237).

3.2 Higher Cognitive Learning

Higher cognitive learning (e.g. abstract concepts) involves more complex
processes than those described by behaviorists such as Pavlov, Watson, Thorndike
and Skinner. Learning a new topic necessitates the creation of mental images and
structures. Some of the mental structures underlining cognitive learning are
thinking, memory, concept formation, problem-solving and intelligence (see week
6 for more information) (Zimbardo, Weber, & Johnson, 2003, pp.237).

17
References
Bandura, A. (1971). Social Learning Theory. General Learning Corporation.

Bandura, A., & Walters, R. H. (1977). Social learning theory

Zimbardo, P. G.,Weber, A. L, & Johnson, R. L. (2003). Psychology (4th ed.).

Boston. USA: Alyn and Bacon.

Blosser, P. E. (1973). Principles of Gestalt psychology and their application to

teaching junior high school science. Science Education, 57(1), 43-53.

Pavlov’s Classical Conditioning Experiment [Online Image]. Retrieved 16

February 2017 from https://psychpics.com/2015/12/11/behaviorism-i-historical-
progress/

Additional Reading
Gormezano, I., Prokasy, W. F., & Thompson, R. F. (1987). Classical conditioning.
Psychology Press. Available at https://books.google.com

McLeod, S. A. (2007). BF Skinner: Operant conditioning. Available at

http://s3.amazonaws.com/academia.edu.documents/44186702/Skinner.pdf?AWSA
ccessKeyId=AKIAIWOWYYGZ2Y53UL3A&Expires=1487241971&Signature=
XcYPybiSC9aAhrF%2B5xhOFf3jOvw%3D&response-content-
disposition=inline%3B%20filename%3DSkinner_-_Operant_Conditioning.pdf

18
Munoz, L. Behaviorist Learning Theory. Available at:
https://lynnmunoz.wordpress.com/learning-theories/behaviorist-learning-theory/

Olson, M. H. (2015). An introduction to theories of learning. Psychology Press.

Available at: https://books.google.com

19