Learning Notes

advertisement
Ivan Pavlov – Russian physiologist (person who studies the workings of the body) who discovered
classical conditioning through his work on digestion in dogs.
Classical conditioning - learning to make a reflex response to a stimulus other than the original, natural
stimulus that normally produces the reflex.
Normally, when food is placed in the mouth of any animal, the salivary glands automatically start
releasing saliva to help with chewing and digestion. This is a normal reflex (involuntary response) in both
animals and humans. The food causes a particular reaction, the salivation. A stimulus can be defined as
any object, event, or experience that causes a response, the reaction of an organism. In the case of
Pavlov’s dogs, the food is the stimulus and salivation is the response.
Unconditioned stimulus (UCS) - a naturally occurring stimulus that leads to an involuntary response.
Unconditioned means “unlearned” or “naturally occurring.”
Unconditioned response (UCR) - an involuntary response to a naturally occurring or unconditioned
stimulus.
Conditioned stimulus (CS) - stimulus that becomes able to produce a learned reflex response by being
paired with the original unconditioned stimulus.
Conditioned means “learned.”
A neutral stimulus can become a conditioned stimulus when paired with an unconditioned stimulus.
Conditioned response (CR) - learned reflex response to a conditioned stimulus.
Sometimes called a conditioned reflex.
Stimulus substitution - original theory in which Pavlov stated that classical conditioning occurred
because the conditioned stimulus became a substitute for the unconditioned stimulus by being paired
closely together.
Cognitive perspective - modern theory in which classical conditioning is seen to occur because the
conditioned stimulus provides information or an expectancy about the coming of the unconditioned
stimulus.
Stimulus generalization - the tendency to respond to a stimulus that is only similar to the original
conditioned stimulus with the conditioned response.
Stimulus discrimination - the tendency to stop making a generalized response to a stimulus that is
similar to the original conditioned stimulus because the similar stimulus is never paired with the
unconditioned stimulus.
Extinction - the disappearance or weakening of a learned response following the removal or absence of
the unconditioned stimulus (in classical conditioning) or the removal of a reinforcer (in operant
conditioning).
Spontaneous recovery - the reappearance of a learned response after extinction has occurred.
Conditioned emotional response (CER) - emotional response that has become classically conditioned to
occur to learned stimuli, such as a fear of dogs or the emotional reaction that occurs when seeing an
attractive person.
CERs may lead to phobias – irrational fear responses.
Vicarious conditioning - classical conditioning of a reflex response or emotion by watching the reaction
of another person.
Conditioned taste aversion - development of a nausea or aversive response to a particular taste
because that taste was followed by a nausea reaction, occurring after only one association.
Biological preparedness - the tendency of animals to learn certain associations, such as taste and
nausea, with only one or few pairings due to the survival value of the learning.
Operant conditioning - the learning of voluntary behavior through the effects of pleasant and
unpleasant consequences to responses.
Thorndike’s Law of Effect - law stating that if a response is followed by a pleasurable consequence, it
will tend to be repeated, and if followed by an unpleasant consequence, it will tend not to be repeated.
Behaviorist; wanted to study only observable, measurable behavior.
Gave “operant conditioning” its name.
Operant - any behavior that is voluntary.
Learning depends on what happens after the response — the consequence.
Reinforcement - any event or stimulus, that when following a response, increases the probability that
the response will occur again.
Primary reinforcer - any reinforcer that is naturally reinforcing by meeting a basic biological need, such
as hunger, thirst, or touch.
Secondary reinforcer - any reinforcer that becomes reinforcing after being paired with a primary
reinforcer, such as praise, tokens, or gold stars.
Positive reinforcement - the reinforcement of a response by the addition or experiencing of a
pleasurable stimulus.
Negative reinforcement - the reinforcement of a response by the removal, escape from, or avoidance of
an unpleasant stimulus.
Punishment - any event or object that, when following a response, makes that response less likely to
happen again.
Punishment by application - the punishment of a response by the addition or experiencing of an
unpleasant stimulus.
Punishment by removal - the punishment of a response by the removal of a pleasurable stimulus.
•
Severe punishment may cause the child (or animal) to avoid the punisher instead of the
behavior being punished, so the child (or animal) learns the wrong response.
•
Severe punishment may encourage lying to avoid the punishment (a kind of negative
reinforcement)—again, not the response that is desired.
•
Severe punishment creates fear and anxiety, emotional responses that do not promote learning.
If the point is to teach something, this kind of consequence isn’t going to help.
•
Hitting provides a successful model for aggression.
•
Punishment should immediately follow the behavior it is meant to punish.
•
Punishment should be consistent.
•
Punishment of the wrong behavior should be paired, whenever possible, with reinforcement of
the right behavior.
•
Punishment should be private
•
Fixed ratio schedule of reinforcement - schedule of reinforcement in which the number of
responses required for reinforcement is always the same.
•
Fixed interval schedule - of reinforcement schedule of reinforcement in which the interval of
time that must pass before reinforcement becomes possible is always the same.
•
Variable interval schedule of reinforcement - schedule of reinforcement in which the interval of
time that must pass before reinforcement becomes possible is different for each trial or event.
•
Ask the students if they can think of some reflexes that might be easy to test in the classroom. If
they do not come up with it on there own, suggest the “eye-blink” reflex. This serves well as an
unconditioned response when the unconditioned stimulus is a puff of air. A word (e.g., the name
of your course, school, etc.) can serve nicely as a neutral stimulus.
•
A discriminative stimulus is any stimulus that provides an organism with a cue for making a
certain response in order to obtain reinforcement.
•
Instinctive drift - tendency for an animal’s behavior to revert to genetically controlled patterns.
•
Each animal comes into the world (and the laboratory) with certain genetically determined
instinctive patterns of behavior already in place.
•
These instincts differ from species to species.
•
There are some responses that simply cannot be trained into an animal regardless of
conditioning.
•
Behavior modification - refers to the application of operant conditioning to bring about such
changes.
•
For example, if a teacher wants to use behavior modification to help a child learn to be more
attentive during the teacher’s lectures, the teacher may do the following:
•
1. Select a target behavior, such as making eye contact with the teacher.
•
2. Choose a reinforcer. This may be a gold star applied to the child’s chart on the wall, for
example.
•
3. Put the plan in action. Every time the child makes eye contact, the teacher gives the child a
gold star. Inappropriate behavior (such as looking out of the window) is not reinforced with gold
stars.
•
4. At the end of the day, the teacher gives the child a special treat or reward for having a certain
number of gold stars. This special reward is decided on ahead of time and discussed with the
child.
•
Behavior modification - the use of operant conditioning techniques to bring about desired
changes in behavior.
•
Token economy - type of behavior modification in which desired behavior is rewarded with
tokens.
•
Time-out - a form of mild punishment by removal in which a misbehaving animal, child, or adult
is placed in a special area away from the attention of others. Essentially, the organism is being
“removed” from any possibility of positive reinforcement in the form of attention.
•
Applied behavior analysis (ABA) – modern term for a form of behavior modification that uses
shaping techniques to mold a desired behavior or response.
•
Biofeedback- the use of feedback about biological conditions to bring involuntary responses
such as blood pressure and relaxation under voluntary control.
•
Neurofeedback - form of biofeedback using brainscanning devices to provide feedback about
brain activity in an effort to modify behavior.
•
Neurofeedback has been used to treat a number of disorders and conditions in the last few
years:
•
• epilepsy
•
• anxiety disorders
•
• depression and anger
•
• drug addiction
•
• chronic fatigue syndrome
•
• obsessive-compulsive disorder
•
• autism
•
• attention-deficit hyperactivity disorder
•
Playing a video game to help solve problems like anxiety and attention problems? It sounds like
a child’s dream world.
•
One of psychologist Edward Tolman’s best-known experiments in learning involved teaching
three groups of rats the same maze, one at a time (Tolman & Honzik, 1930). In the first group,
each rat was placed in the maze and rewarded with food for making its way out the other side.
The rat was then placed back in the maze, rewarded, and so on until the rat could successfully
solve the maze with no errors—the typical maze-learning experience.
•
Edward Tolman’s best-known experiments in learning involved teaching three groups of rats the
same maze, one at a time (Tolman & Honzik, 1930).
•
Group 1 – rewarded each time at end of maze.
•
Learned maze quickly.
•
Group 2 – in maze every day; only rewarded on 10th day.
•
Demonstrated learning of maze almost immediately after receiving reward.
•
Group 3 – never rewarded.
•
Did not learn maze well.
•
Latent learning - learning that remains hidden until its application becomes useful.
•
Why would the conditioned dogs refuse to move when shocked? The dogs that had been
harnessed while being conditioned had apparently learned in the original tone/shock situation
that there was nothing they could do to escape the shock. So when placed in a situation in
which escape was possible, the dogs still did nothing because they had learned to be “helpless.”
They believed they could not escape, so they did not try.
•
In one of his more famous studies (Köhler, 1925), he set up a problem for one of the
chimpanzees. Sultan the chimp was faced with the problem of how to get to a
•
banana that was placed just out of his reach outside his cage. Sultan solved this problem
relatively easily, first trying to reach through the bars with his arm, then using a stick that was
lying in the cage to rake the banana into the cage. As chimpanzees are natural tool users, this
behavior is not surprising and is still nothing more than simple trial-and-error learning.
•
But then the problem was made more difficult. The banana was placed just out of reach of
Sultan’s extended arm with the stick in his hand. At this point there were two sticks lying around
in the cage, which could be fitted together to make a single pole that would be long enough to
reach the banana. Sultan first tried one stick, then the other (simple trial and error). After about
an hour of trying, Sultan seemed to have a sudden flash of inspiration. He pushed one stick out
of the cage as far as it would go toward the banana and then pushed the other stick behind the
first one. Of course, when he tried to draw the sticks back, only the one in his hand came. He
jumped up and down and was very excited, and when Köhler gave him the second stick, he sat
on the floor of the cage and looked at them carefully. He then fitted one stick into the other and
retrieved his banana.
•
Observational learning - learning new behavior by watching a model perform that behavior.
•
Learning/performance distinction - referring to the observation that learning can take place
without actual performance of the learned behavior.
•
ATTENTION - To learn anything through observation, the learner must first pay attention to the
model.
•
MEMORY - The learner must also be able to retain the memory of what was done, such as
remembering the steps in preparing a dish that was first seen on a cooking show.
•
IMITATION - The learner must be capable of reproducing, or imitating, the actions of the model.
•
MOTIVATION - Finally, the learner must have the desire to perform the action.
•
Directions for activity: This activity takes about 5 to 10 minutes. Ask for a volunteer from the
class. Tell everyone that the class is going to attempt shape the volunteer’s behavior using strict
principles of reinforcement. Escort the volunteer out of the room where he/she will remain for a
few minutes. Return to class and decide on a behavior (e.g., something relatively simple like
turning off a light or standing up on a table) that the volunteer would have a good chance of
performing with class applause as the only source of feedback.
•
Inform the class that no verbal instructions are to be given to the volunteer, but that they
should applaud more vigorously as the volunteer moves closer to completing the simple
behavior. It’s kind of like the old “hot and cold” game that many of us played as children.
Remind the class of the rules for shaping and then bring the volunteer back into the room to see
if shaping really works.
•
Following the activity, discuss the value of both verbal feedback and reinforcement in the
process of shaping and learning.
•
Download