Download Week 3 - Stephen P. van Vlack

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Applied behavior analysis wikipedia , lookup

Neuroeconomics wikipedia , lookup

Verbal Behavior wikipedia , lookup

Behavior analysis of child development wikipedia , lookup

Insufficient justification wikipedia , lookup

Behaviorism wikipedia , lookup

Learning theory (education) wikipedia , lookup

Eyeblink conditioning wikipedia , lookup

Learning wikipedia , lookup

Psychophysics wikipedia , lookup

Psychological behaviorism wikipedia , lookup

Classical conditioning wikipedia , lookup

Operant conditioning wikipedia , lookup

Transcript
Stephen van Vlack
Sookmyung Women’s University
Graduate School of TESOL
Human Learning and Cognition
Spring 2015
Week 3 – Answers
Lieberman (2004) Chapter 2 Foundations of classical conditioning
1. What is classical conditioning and why is this important?
Classical conditioning, otherwise known as Pavlovian conditioning, is when an organism
forms a simple association between two previously unrelated stimuli. This association is
claimed to be based on, or at least strongly affected by, contiguity.
To understand this we need to back up a bit and take a look at some more basic ideas.
These are the ideas of:
Stimulus – like input itself, this is something that we can expect people to react to
Response – the reaction to the stimulus/input
Habituation – when people get used to a stimulus overtime and fail to react to it
Learning – when a response is seen to occur in reaction to a stimulus and this is not seen
to occur in the natural world. An organism has made an association between
two things on the basis of experience. That is the association is learned (not
innate or natural)
Non-learning - a situation in which things do happen to co-occur (are contiguous) in the
experience of the organism do not become associated. There is no change in
the organism’s behavior despite what occurs or is contiguous in the
environment.
Pavlovian conditioning is important in the study of learning because it shows that
organisms can LEARN to form associations between two things that are not causally
connected in the world and these associations are discernable though the behavior of the
organism. We start by looking at the different types of stimuli and the responses that can
be generated from them.
One of the stimuli is important to the organism and is called one of two names. The
general name for this type of response is an orienting stimulus. If this same orienting
stimulus is considered from the perspective of Pavlovian conditioning it is called an
unconditioned stimulus (US). This type of stimulus is often though of as being
evolutionary or natural because this type of stimulus evokes a response that may keep a
person or animal alive in a dangerous situation. For example, if the wind blows in your
eyes you will react by squinting/closing your eyes or turning your face away from the wind
(to avoid getting dust in your eyes). So the wind is an orienting or unconditioned stimulus.
The response to an US (in this case the wind) is seen as being natural and thus not
learned from the perspective of classical conditioning. It is called an orienting response or
an Unconditioned Response (UR) and it is not learned. It is simply natural based on our
physiology.
The other type of stimulus is called a conditioned stimulus (CS). The association made
between a conditioned stimulus and a response, on this view, must be learned, as there is
a demonstrable change in behavior as a result of an association having been made. For
example, a startle reaction to a rose plant is seen as being a conditioned response to a
conditioned stimulus because in the natural world this type of stimulus does not always
lead to this type of response. The startle reaction must have been learned somewhere or
somehow.
In Pavlovian conditioning the association of a CS to a CR is made possible by US. The
organism will have transferred the natural or unconditioned response to the
unconditioned stimulus to what is now seen as a conditioned stimulus even though prior
to the experimental trials using temporally controlled exposure they were not at all
related. In a nutshell, this is what classical conditioning is. Again, it is used to show that
responses can be learned and as a response to what is found incidentally in the
environment.
2. What is extinction and how does it work?
Extinction is an important step in understanding learning because it shows us that
learning is not just a result of a simple association being made between two contiguous
things. Learning is not static - it changes. Extinction, then, is the eradication of the
conditioned response due to breaking the co-occurrence of the unconditioned stimulus
and the conditioned stimulus. An interesting thing about extinction (and this gives us
valuable insight on learning and forgetting in general) is that the effects of extinction are
only temporary so long as the pattern of co-occurrence is reinstated at a later date. This
confirms what we said last week about the theory that once learned things are never
completely forgotten or lost though they might go seemingly dormant for a period of
time (for a number of reasons). In any case, the potential for recovery is always there.
What probably actually happens in an extinction situation is that the costs of retaining
or even recovering the association simply becomes comes too high. This is based on the
idea of neuronal firing. We know that neuronal firing is a process in the brain by which
electricity is entered into a neuron and is disseminated to its synapses. Each synapse
requires a different amount of electricity to fire it, or make it work. The amount of
electricity required is called the threshold level. As a synapse is used more and more
often its threshold level will go down. A low threshold level means the nodes at the end
of the neuron’s axon fire more easily (using less electricity). This also leads to higher
activation level – which also means the neuron will fire with less effort. So, to a large
extent the threshold levels of a neuron’s synapses are determined by their frequency. If
synapses are not used frequently their threshold levels are going to go up and the longer
they are not used the more they will go up and the more energy it will take to fire those
synapses. Once they reach a certain level it becomes hardly unlikely that these synapses
will be fired because they simply cost too much energy. The brain due to its plasticity will
simply find another way around and threshold levels will therefore continue to go up in
what seems to be a rather vicious cycle. This doesn't mean we can't fire a synapse that
has a high threshold level due to lack of use, it just means that we avoid doing it because
it's not efficient. If we need at some point to actually do that we can but it is going to cost
a large amount of energy
3. How is conditioning related to learning?
Classical conditioning as mentioned in the answer to question 1 is a simple example of
what is called associative learning. This is learning where one learns to associate two
previously unrelated concepts, ideas, and stimuli. It should be clear that we are
expanding out from the simple stimulus-response (s-r) model mentioned in Lieberman
(2004) and are importantly including a stimulus-stimulus (s-s) model here. The key
difference between the two is that the s-r model includes some sort of physical reaction
(response) to a stimulus. The s-s model takes us inside the brain and allows us, by
extending out from the simplistic s-r model, to posit associations that are not just physical
but cognitive. In short, the s-s model allows us to posit associations between concepts
and this is something we need if we are to extend the basic model of classical
conditioning to the learning of systems like language.
Although for the purposes of trying to control the process and/or concepts of morality
classical conditioning experiments are usually carried out in a laboratory and with
organisms other than humans, it is possible to see how this can be applied to the outside
world and to the human world in general. Conditioning is a simple type of learning that
takes natural association just one baby step further. In conditioning we are taking what
was learned through an UR and extending it on to another stimulus that just so happens
to co-occur (is contingent) in the world. Such a type of learning is obviously largely
responsible for our uncanny ability to survive for it means that we can extend our
behaviors beyond a simple flight or fight response initiated by the association of an US
and an UR. It may also be the very underpinning of our ability for reasoning and possibly
even language as we shall see.
4. How can classical conditioning be used to explain some language learning
phenomenon?
Classical conditioning as mentioned in the answer to question 1 is a simple example of
what is called associative learning. This is learning where one learns to associate two
previously unrelated concepts, ideas, and stimuli. It should be obvious that we are
following the stimulus-stimulus (s-s) model here. As mentioned above, it is the leap from a
s-r model to a s-s model that allows us to bridge into language learning.
The very defining feature of language is its arbitrariness as proposed by de Saussure
(1959); that is the arbitrariness of linguistic structures to their prescribed or adopted
meaning. We would then wonder, if there is no logical reason why some structural units
should be expected to be found adjacent to others (to think so would be ludicrous) then
how did they come to co-occur like they do? How are word forms associated with
meanings? How can a person learn these patterns of association? There must be a type of
learning which supports the rather quick learning of language. Association based on
contiguity might just be the missing link. Think about context. Context is the clue to all
learning and those that can see it and differentiate its parts are the good learners.
Lieberman (2004) Chapter 3 Principles and applications of classical conditioning
1. How does contiguity affect conditioning?
Contiguity refers to the timing of the conditioning. This relates to the intervals between
the presentation of the two stimuli as well as the order of the stimuli. In looking at
contiguity the biggest lesson we need to take from this is that a forward sequence is the
most effective. This probably because a forward sequence would allow the organism to
effectively prepare for the coming unconditioned stimulus (anticipatory priming). It is
also very important to remember that contiguity as a variable in conditioning is very much
reliant on the task itself, the species involved in the experiment, and the type of response
expected. All this ties into brain differences among species and among the different
functions as related to the type of response expected. Environmental differences have an
effect. For humans it is also cultural differences (prior knowledge) that affect the way we
evaluate stimuli. If the stimulus is primary visual in nature then the neurological structure
of the optic nerve and how this is connected to other parts of the brain becomes an
important factor. Also, some experiences are more intense than others, such as food
poisoning, and this will have a large effect on contiguity.
There are several other factors that have a strong effect on conditioning within and
despite contuguity. They are:
Prior exposure - familiarity,
Compound stimuli,
Surprise,
Relevance,
and Inhibition.
Prior exposure is a sword that cuts both ways. Prior exposure in which the two stimuli did
co-occur wills serve to reinforce or heighten the conditioned response. Prior exposure,
however, in which the two stimuli did not co-occur will increase the chances that a
conditioned response will not occur.
When compound conditioned stimuli are presented several different effects occur
reaction depending on the timing and salience of the conditioned stimulus in relation the
unconditioned stimulus. Most of this is intuitive and, therefore, does not require further
explanation here, as are the effects of surprise and blocking when two or more
conditioned stimuli are used.
The amount of relevancy of the two stimulus to each other seems to have a fairly large
effect on not only what will be associated but also on how quickly the association will take
place.
An inhibitory response is one in which the absence of the unconditioned response will
cause a conditioned response. What is interesting for us is that an inhibitory response can
only be generated based on an understanding (knowledge) of the components that go
into making the unconditioned stimulus. In order to know that something will not happen
you need to be able to predict under what conditions it would happen and to know about
what elements are present and what elements are missing.
2. What is instrumental learning and how does it differ from classical conditioning?
Instrumental learning is a kin to the idea of preparedness as presented in Lieberman
(2004).
Instrumental learning is learning in which a stimulus, response, and the consequences of
the action derived from the first two are associated. This is a based on the simple and
necessary extension of what we observed in classical conditioning to the real world of
consequences and rewards. In instrumental learning the outcome of the newly acquired
behavior (response) is evaluated against a series (possibly) of needs-based criteria. In this
way the consequences have some effect on the association between the first two
elements in the triumvirate (stimulus – response - consequences). While this new
dimension brings behavior theory closer to the real world of cause and effect
relationships (trial and error learning) it also adds a further dimension of complications
and differing variables to the behavior-based model.
The beauty of instrumental learning, and one of the main differences between the
instrumental learning and classical conditioning, is that of control. In instrumental learning
because the consequences are taken into consideration the organism is able to control
actions consciously. This means that once someone has made two types of associations
which are necessary in instrumental learning they will/may actually seek out situations and
engage the world in order to receive a consequence which they would like. This means
that they need to plan. In classical conditioning, on the other hand, the organism is really
just a passive sponge-like creature which observes what happens in the world and
manages to associate things which happened to co-occur. There is no volition or
planning, just a response.
We may also take a moment here to make a connection between instrumental learning
and the concepts in socio-cultural theory of artifacts and tools. In fact, it seems that the
learning of artifacts as tools within sociocultural theory is a direct result of instrumental
learning. In essence, we learn what we can do with different things by examining and
evaluating the effect/consequences of the response. In this way we learn the
repercussions of our actions.
3. What is response learning and what is it importance?
The basic idea in response learning is that the specific stimulus is cut out (removed) and
response exists unto itself and without a stimulus. In effect, what basically happens in
response learning is that the consequence becomes the stimulus to a large extent. For
psychologists, response learning is quite important because it allows them to shape
responses and in doing so they can actually alter people's behavior particularly when
there are problems which need to be fixed in behavior. For us, however, response
learning is interesting because it ties in to language itself. Certainly in the real world when
people use language they don't have to wait for a particular linguistic stimulus in order to
start talking. If this were the case then nothing would ever get done because nobody
would actually talk first. So, for us to develop a theory of language learning that is based
on basic learning concepts of neuronal firing helped by behaviorist models then response
learning becomes quite important. Again, the bottom line in response learning is that
people will actually respond without a stimulus. So, for language this would mean people
actually start speaking without having to wait for some sort of linguistic stimulus. People
do this because they think they know what the consequence of their speech is going to
be. This is, therefore extremely, basic and is extremely important for us.
As was mentioned above, what is learned is the connection between not only the stimulus
and the response as we saw in classical conditioning but both of them with the
consequence. This is demonstrated by changes in behavior relative to the type and
strength of the consequence. What it really interesting here is that this three-way
connection can be broken. Thus, the stimulus might not be important all the time for
response to occur. Likewise people can respond without being assured of a particular
consequence, so the consequence will eventually be cut off from the whole scheme. By
advocating this three-way type of connection or association, but one which is also
flexible, we can allow for a tremendous amount of variation in behavior based on simple
learning. This is the type of thing which is necessary for us if we're really going to use this
to describe how language might be learned and actually works in the real world.
4. How do the contingencies of non-reward, punishment, and avoidance work in
instrumental conditioning?
We have already dealt with instrumental learning which is basically all about trial and error
learning in making associations between three different elements; the stimulus, response,
and consequence. Now, we are ready to start looking at us how we can use this
information to condition people (our students) to behave in the way that we want. This
means that we can use this knowledge to reinforce behavior that we like, such as studying
and paying attention in class, but we can also use this information to try to alter behavior
that we do not like, such as running around in class and not paying attention. In addition,
there are also certain linguistic elements or language learning elements which we can also
further define based on the type of information contained in theories of instrumental
learning. This week in particular we are going to focus on trying to alter unwanted
behaviors and reinforce better behaviors after having reprogrammed the unwanted
behavior. This is done in response to what we have already discussed which is basically
about reinforcing good or wanted behavior.
The three main contingencies available to us in the theory of instrumental conditioning in
order to try to change behavior are non-reward, punishment, and avoidance. Starting with
the idea of non-reward we see that it is actually a very simple idea. All this really entails is
simply taking the reward away (or really not providing a reward) for a particular type of
behavior. The expected effect would be that the behavior would stop. For example, I give
you stickers on your homework assignments basically as a type of reward for doing the
work. This is a reward because even if you don't do a great job you still get some sort of
sticker. If I were to stop giving you stickers based on your performance on the homework
your behavior would probably change. You might be less motivated and your drive to get
a sticker would diminish because, hey!, there was no sticker. This seems very simple and
straightforward but in the real world it might not actually be so easy. To start with we
would need to know what the specific rewards are as a result of a particular behavior and
try to remove them. If the reward is coming from somewhere that's out of our control like
some sort of internal reward then there is actually very little we can do. Research on lab
animals has also shown that even when the reward is removed they still seem to perform
the action because they are used to it. It's a type of habituation where you get used to
performing certain actions regardless of whether the reward is there or not.
Punishment is about trying to stop a certain behavior by changing the consequence from
a positive one to a decidedly negative one. For example if a who child acts up in class is
somehow rewarded for this by getting attention from other students, by getting universal
laughter and an accompanying feeling of power then the teacher needs to find a
punishment which is stronger than the positive consequences. It has been claimed, most
notably by Skinner and Thorndike, that punishment doesn't work, at least not long-term
but there have been many studies which show that punishment does work provided it is
given immediately, consistently, and with a strong force (brutally, so as to eclipse any
possible reward). What I mean by brutally is that it has to be a very strong punishment
immediately and always. If the punishment starts off small and get stronger then the
person will simply habituate to the punishment and it won't have the same effect.
Although punishment does seem to work in some ways there is also a moral dilemma in
punishment. For punishment to work it must be brutal and strong. This often has the
effect of causing aggression among the punished group or person. Also punishment can
also be sending mixed signals in that people still might be getting rewards as their
punishment and some possible positive rewards become associated together. This was
Skinner's main argument. It is, therefore, still important that the punishment really does
strongly outweigh any kind of possible reward. Since we are not always available to
punish students in every instance of behavior and, therefore, cannot control all the
rewards this becomes somewhat of a tricky venture.
In avoidance, the third type of contingency, the behavior itself changes in order to avoid
any kind of consequence. In order for this to happen it would seem to reason of course
that the consequence would have to be negative. So, if we don't want something bad to
happen we do something in advance which will stop that negative consequence. This
means that the behavior itself which would cause the consequence doesn't exist (is simply
avoided) and is replaced by another type of behavior. As we said in class, the typical type
of avoidance behavior that we find in the English language classroom in Korea would
simply be silence. Students are silent because they are avoiding the punishment internal
and external that comes from making mistakes. This punishment could and probably did
originally come from the outside but it also comes from the inside with people that are
highly sensitized to it. This type of silence in behavior probably is also a very strong result
of a feeling of helplessness as we will discuss in the next question.
5. What are some of the main applications of these ideas for us as language teachers?
Punishment
Motivation
There are many applications of these ideas in instrumental learning which we can apply to
the language classroom. Most of these applications relate to the idea of motivation or to
put it into more behaviorist terms drive and incentive. We can regulate drive and
incentive by carefully trying to regulate punishment and reward using the principles
presented here. Based on some of the ideas presented here it would seem that a certain
amount of reward is necessary in order to give people drive, but at the same time we
don't want this drive to become satiated through an overuse of rewards. When the drive
does become satiated then people stop working so hard and drive reduces. That is the
problem when we provide too much of what is, in our familiar terminology, extrinsic
motivation. Therefore it's important to know when and how to reward and when and how
to punish. The danger of punishment is the people begin to associate the punishment
with the response and will therefore avoid responding in all possible cases and I think we
see things like this actually happen in the English language classroom in Korea. The
evaluation system is a system of punishment not a system of reward for the vast majority
and people do associate punishment with responses, in large part because their
responses are never quite good enough. What this means for us is that we need to
reward but not for reward's sake. We need to reward honestly and fairly and when it's
applicable. Punishment should be dished out the same way. There should be no
irrelevant patterns of punishment or reward.
An underlying presumption of this idea is that the students are actually responding. This
means that the students actually have to perform (respond) in some sort of way. We can
only get an idea of consequences or form associations with consequences when we
actually respond. This would seem to indicate that language classrooms must involve a lot
more real language use for the students. If language classrooms do not involve actual
performance, that is responses, by the students then none of these ideas are going to
work.
As we move further and further into these behaviorist theories it is important to
remember how they might actually be linked to language use and eventually language
learning. We are interested in language as a specific type of behavior. This will involve
language as being in its primary sense; a type of behavior. Of course, for this behavior to
actually occur there has to be some underlying knowledge and language is a very
complex system of knowledge. This is what we're trying to work towards. The things we
study this week about instrumental learning are quite important because now we not only
can explain to how patterns of sounds, morphemes, words might actually enter the brain
but now we can actually begin to talk a little bit about behavior; how speakers use them.
In instrumental learning what people actually learn is to associate consequences with
either stimulus or responses or both. Knowing these consequences allows people to selfdirect behavior. This means that language, for our considerations in this course, has now
moved out of the realm of the brain and into the realm of real-world behavior. People can
begin conversations and engage in conversations based on somebody's very simple ideas
here. The fact, mentioned above, that this three-way association between stimulus
response in consequence can be broken in certain places allows people to extend the
simple generalization into many different areas, thus allowing language to actually occur
as a type of behavior. This is very interesting in importance step for us in our quest to try
to determine how language is actually learned.
References
de Saussure, F. (1959). Course in general linguistics. New York: McGraw-Hill.