In the last section, we
described the variety of aspects of reinforcement and how that related to
operand conditioning. In this section we begin by discussing some concepts
that are related to reinforcement and so let’s begin with a discussion of
the first concept, and that is discrimination and that’s shown in slide two.
Discrimination that we talk about in operand conditioning
is highly similar to the ones that are in classical conditioning, it’s just
a little minor tweak, that is, responses which are reinforced in one
situation and not in another will reoccur only in the situations where they
were reinforced, so and the classic example goes along with that, is in
slide three. For example, in a bar, loud talking can be reinforced and you
get lots of attention from that, but in the classroom, loud talking is not
reinforced and you may in fact get you kicked out of class.
The second example relates to the little kid that we
talked about in the last section who was yelling in the store for a candy
bar, so the kid may be reinforced for yelling in the store, mom gives it the
candy bar or dad gives it the candy bar, but when it’s at home and it starts
yelling and screaming I want this and that and everything else, other things
may occur instead, that is some kind of punishing procedure. So the kid
learned very, very quickly that it’s ok for it to yell in the store and it
will get a reinforcer, but when it’s at home, other things may occur, ala
spankings or whatever, depending on what your punishment procedures are that
you use.
Now second concepts that’s related to reinforcement is
generalization and this is the same kind of generalization that we also
talked about in classical conditioning and again, it’s shown in slide four
and that is new stimuli that are similar to the original stimuli are
reinforced and basically that occurs when an organism is reinforced for
other responses that are similar to the original response. So for example if
you are reinforced for yelling, you may talk loudly and get reinforced, ok,
so in essence instead of yelling you talk loudly or when you get
reinforcement under some conditions, ala, when you wear certain clothes and
you wear similar clothes you also get reinforced. So again, in each of these
situations, generalization occurs. The organism basically is reinforced when
stimuli are similar to the original conditioned stimuli.
Now in addition to aspects of generalization and
discrimination, let’s talk about some other aspects that relate to
reinforcement and the first one of these relate to types of reinforcers and
that is shown in slide five. Primary reinforcers are reinforcers that innate
reinforcing. These are things such as food, sex, water, etc., in essence
they are the things that are genetically reinforcing to you.
These are in contrast to secondary reinforcers that we
talk about in slide six. In secondary reinforcers, these are reinforcers
that acquire their ability to reinforce through experience. The classic
example is money and usually these secondary reinforcers become reinforcing
when they can be exchanged for primary reinforcers. For example, we can
exchange money for food or we can exchange money for other things. But
secondary reinforcers don’t just have to be money, they can be other things.
They can be stars, they can be tokens and many, many, many other things, so
in essence, what we have and most of our society basically operates on
secondary reinforcers but we can exchange in essence for primary reinforcers.
So what are some variables that can influence some
secondary reinforcers and we talk about that beginning in slide seven. And
the first thing is the amount of primary reinforcer that is given, that is,
the more primary reinforcement that is given with the secondary, the
stronger the secondary reinforcer becomes, so if you get lots of food when
you provide some small amount of money, that becomes extreme, that secondary
reinforcer becomes extremely powerful.
Number two, the number of pairings between the primary
reinforcer and the secondary reinforcer. In essence, the more pairings that
you have, the stronger the secondary reinforcer becomes. And this continues
on in slide eight.
Number three, the time between the presentation of the
primary and secondary reinforcer and that is, the more time, that is the
time delay between the primary reinforcer and the secondary reinforcer, the
poorer the secondary reinforcer becomes. So in essence what you wanta do is
have a very, very short time between when that secondary reinforcer and that
primary reinforcer occur together.
Now there’s a wide variety of variables that influence the
rate of conditioning and so let’s begin by talking about what variables
these are and how it works and so we begin doing that starting in slide 10.
And the first one relates to the quantity of reinforcers
given. The more reinforcement, in essence, you give, the faster the organism
will respond. So but the problem can become is if you get too much food, the
organism gets satiated, so for example if you’re using food or food pellets,
the organism gets full after a while, so you have to kind of be careful
about what you giving.
The second variable relates, as we see here in slide 11,
to the quality of the reinforcer. The better the quality of the reinforcer,
the higher the rate of the responding and the faster the responding will
occur. So let’s say that you are reinforcing your dog. Well if you give
Alpo, that’s basically better than giving something out of a bag that you
might buy at some local grocery store and so the animal will respond much
faster to the Alpo than it will at some kind of cereal based doggy food. In
addition to that, the quality is better depending upon the organisms. For
example, for myself, chocolate chip cookies are significantly better
reinforcers for me than oatmeal wafers because there are no other things
than chocolate chip cookies.
The next variable that influences the rate of conditioning
relates to the time delay between the response and the reinforcer and this
is shown in slide 12. Reinforcers are more effective the sooner they are
presented following the response and in general, the longer the delay, the
lower the response rate, so what you, when the organism makes a response,
you want to try to immediately reinforce it for whatever that particular
behavior you are trying to get increased. So for example, if you want your
kid to do something, and do it more often, as soon as it makes that
particular response, you want to immediately give it the reinforcer that
you’re using. Again, the longer the delay, the lower the response rate, so
if a kid does something really good, lets say it goes out and cleans the
yard, what you should do is if you want more yard cleaning at a faster,
higher rate, you should immediately go out and give it $5, instead of
waiting until the end of the week.
The rate of reinforcement will also have a major
importance within the time period as well and that’s also shown in slide 12.
That is, the number of reinforcers that are given in some particular time
period and in essence the smaller the time period between reinforcers, the
more responding that will in essence occur, so if you wanta get high
responding you want to basically give a large number of reinforcers of high
quality and have very minimal time between the time the organism makes the
response and the reinforcers received.
The next variable and shown in slide 13, relates to a
concept that we’ve already discussed and that relates to the concept of
deprivation. Deprivation is extremely important in working with behavior and
the longer the time period since the last reinforcer was consumed, the
higher in essence the response rate and an animal in essence responds faster
the more deprived it is, so if a person doesn’t have any money, it will
respond very, very rapidly and very highly when it has minimal amounts of
money and that can occur with the rate of reinforcers or be independent of
it.
Now the next variable that’s going to influence the
reinforcement procedures relates to the percent of reinforcers that are
contingent upon a behavior and this is shown in slide 14. Now let’s just
make a basic concept and understand something. You don’t need to have a
person respond to give a reinforcer and oftentimes in our society, we just
give reinforcers willy nilly. That is, I just see my kid and I give him 5
bucks to let him do whatever he wants. And so in essence on the other hand,
I could make my kid do only work to in essence get some money and that would
in essence be where he’s only working for me and then he would have,
everything is based on work. So in essence we break that down into
contingency and that is the percentage of reinforcers that are contingent
upon a behavior. When there is no contingencies, that is, you have nothing
that’s out there that’s required and everything that’s free basically the
organism gets the reinforcer whether or not a response is made and that’s
the classic example of the welfare state. You don’t have to do anything, you
just get your $5,000 every month.
On the other hand, you could have a concept where
everything is 100% contingent, that is all the enforcers must be obtained by
making some kind of a response, that is, you have to work for a living to
get the particular reinforcers that you have.
Now the next concept relates as we see in slide 15,
relates to the context in which the reinforcement occurs. And in essence as,
and this kind of goes back to some of the (can’t understand) models that we
talked about earlier. In essence the more reinforcers that are available,
the less effective any one particular reinforcer will be and for example,
food, money, love, etc., are all available, if those are all freely
available, which of those becomes very, very important for you that you
would do things for.
Now we’ve talked about a variety of different concepts
that relate to operand conditioning. In this section right now, what we’re
gonna start talking about is a little bit of applications or relationships
within reinforcement. And so the first kind of concept that we wanta talk
about is related to what I call discriminative stimuli and related notions.
When stimuli are signaled that some kind of behavior will
be reinforced, as we see in slide 16, they are called discriminative
stimuli, or what is called an SD. In contrast, there are stimuli that signal
non-reinforcement and these are what are called S deltas. And a related
concept of both of these SDs and S deltas relates to what we call
differential reinforcement. And that relates to basically reinforcing a
response in the presence of one stimulus and not reinforcing the same
response in the presence of another stimulus.
Let’s take an example of door knocking. Right now on the
outside of my, of the door here where I’m doing my, this taping is a sign
and it says do not knock and there is little dart that’s in it. Now what
would happen if I, if somebody knocks on the door, all right, and if the
door was shut and that do not knock is on the door, I would not in essence
reinforce the person by going over and answering the door. Now if they
persisted in knocking, what would happen. I would stop, I would get up and I
would be very upset with the person who’s been knocking on the door when
there’s a big sign that says do not knock. So in this case, the sign becomes
in essence the stimulus where I will not reinforce the knocking if the door
is closed and the sign is up, but on the other hand if the door is closed
and the sign is not up, knocking will be reinforced and I will in essence
get up and go over and answer the door. So again, particular stimuli,
basically relate to if you reinforce a response in the presence of one
stimulus and not reinforcing them in the presence of another stimulus and
that relates again to the concept of differential reinforcement.
Now the next aspect of reinforcement relates to
generalization and we’ve talked about generalization before in earlier
sections of this, of operand conditioning and within classical and
instrumental conditioning as well. So what do we talk about in relation to
operand conditioning and this is shown in slide 18. For generalization,
they’re in essence two types—there’s stimulus generalization and response
generalization, so let’s talk about stimulus generalization first. In
essence, stimulus generalization refers to the transfer of a response
through situation other than where the training actually took place. That
is, several stimuli can in essence cause the same response, thus somebody
knocking lightly on the door can cause me to get up and open the door or
tapping on the door or even pounding on the door will cause me to get up and
go over and ala open up the door. So in each of these cases, although
there’s a variety of different types of stimuli, I get up and answer the
door.
Now this is a little different than we see in slide 20
with response generalization. And that is, and this is the one that you kind
of know sort of already and that is the reinforcement of one response
increases the probability of other responses that are closely linked to the
original response. For example, if you go out and kiss someone when they
smile at you or something like that, other people when they smile at you may
also get kisses as well, that’s provided you know these people and it’s ok
to kiss them. On the other hand if somebody, if you kissed one person and
that was ok after they smiled at you, and you kissed another person and they
smacked you, you would know that that would be either a discriminative
stimuli, ala different people or that one person did not like to be kissed
and in essence that would be a punishment procedure which we’ll talk about
in the next section.
So for example we can have a more academic type of thing,
such as the reinforcement basically of one response, ala you’d get an A in
learning, studying for learning and getting an A on your exam also increases
your studying for cognition which ultimately gives you an A as well.
Finally within this section, we have the concept of
chaining. Now chaining’s a little different than what we talked about
earlier with Stuart Little the mouse, but in essence what it is, it has some
overlap and here the organism must be knit of chain of behaviors before they
get reinforced. And as we often saw, as we saw in the Stuart Little example,
we often use reverse shaping to get started. Now chaining doesn’t
necessarily have to be with animals. We all admit a wide variety of
different chains of behaviors before we get reinforced. We do varieties of
different types of work, we do varieties of different types of contact
comfort, etc., with significant others, etc. So in essence all of these
things can give reinforcements but we must do a wide variety of different
behaviors in essence to get the final reinforcement.
So in this section, we had a review of some different
aspects of variables that relate to operand conditioning. In the next
section we will begin talking about other aspects of operand conditioning.
Back