peterson, n. (1978) - an introduction to verbal behavior (1)
DESCRIPTION
verbal behaviorTRANSCRIPT
0
Peterson, N. (1978).
An Introduction to
Verbal Behavior.
Copyright 1978/2002
Norman Peterson,
used with
permission.
1
CHAPTER 1: INTRODUCTORY CONCEPTS ...................................................................................... 3
1-1 MEDIATED REINFORCEMENT .......................................................................................... 3
1-2 VERBAL BEHAVIOR .......................................................................................................... 6
TYPES OF CONTROLLING VARIABLES ..................................................................................... 8
1-3 VERBAL STIMULI .............................................................................................................. 8
1-4 ESTABLISHING OPERATION ........................................................................................... 11
1-5 POINT-TO-POINT CORRESPONDENCE ........................................................................... 14
1-6 RESPONSE-PRODUCT .................................................................................................... 16
1-7 FORMAL SIMILARITY ..................................................................................................... 18
1-8 FORMAL AND THEMATIC CONTROL .............................................................................. 20
CHAPTER 2: ELEMENTARY VERBAL RELATIONS........................................................................... 22
2-1 ECHOIC VERBAL RELATION ............................................................................................ 23
2-2 COPYING A TEXT ............................................................................................................ 25
2-3 TAKING DICTATION ....................................................................................................... 27
2-4 TEXTUAL BEHAVIOR ...................................................................................................... 29
2-5 THE INTRAVERBAL RELATION ........................................................................................ 32
2-6 THE TACT RELATION ...................................................................................................... 35
2-7 THE MAND ..................................................................................................................... 38
A COMPARISON OF THE MAND AND THE TACT .................................................................. 42
2-8 THE AUDIENCE .............................................................................................................. 43
CHAPTER 3: EXTENSIONS OF VERBAL BEHAVIOR ........................................................................ 47
STIMULUS FEATURES........................................................................................................... 48
3-1 EXTENDED TACTS .......................................................................................................... 49
TYPES OF EXTENDED TACTS ................................................................................................ 53
3-2 GENERIC EXTENSION ..................................................................................................... 54
3-3 METAPHORICAL EXTENSION ......................................................................................... 56
3-4 METONYMICAL EXTENSION .......................................................................................... 59
3-5 WAYS OF TEACHING RESPONSES TO PRIVATE STIMULI ................................................ 63
COLLATERAL RESPONSE ...................................................................................................... 65
3-6 MORE WAYS OF TEACHING RESPONSES TO PRIVATE STIMULI ..................................... 68
(3) COMMON PROPERTIES and (4) RESPONSE REDUCTION ............................................... 68
CHAPTER 4: MULTIPLE CAUSATION ............................................................................................ 72
4-1 MULTIPLE CONTROLLING VARIABLES ........................................................................... 72
MULTIPLE RESPONSES ......................................................................................................... 74
2
4-2 FRAGMENTARY SOURCES OF SUPPLEMENTARY STRENGTH......................................... 76
IRRELEVANT ................................................................................................................................. 76
4-3 SUPPLEMENTARY STIMULATION .................................................................................. 79
4-4 PROMPT/PROBE ............................................................................................................ 82
3
CHAPTER 1: INTRODUCTORY CONCEPTS
This text is concerned with the concepts presented in B. F. Skinner’s book titled Verbal
Behavior. Because Skinner’s book was published over a half-century ago, it does not include all
that we now know, or think we know, about verbal behavior. But Skinner’s work is the
foundation for later developments and it also provides a conceptual framework that
practitioners can use when their goal is to teach or improve speaker and listener repertoires.
This text, the creation of Norman Peterson, has been revised and updated with his permission
by Sigrid Glenn for use in this Internet course.
Language scholars have developed many definitions of language based upon what
appear to be its unique features, but there is considerable disagreement on what those
features are. One approach not taken by linguists was that of identifying those variables that
were responsible for the behavior of speaking (or listening). Indeed, verbal behavior was
studied on the basis of form or structure, quite apart from the circumstances under which it
occurred. Skinner’s approach was an attempt to analyze exactly those circumstances under
which verbal behavior occurs and how it is maintained.
Skinner began with the basic assumption that verbal behavior is behavior that is subject
to the same principles that describe the relations between nonverbal behavior and the
environment. In fact, there is only one feature of verbal behavior that sets it significantly apart
from nonverbal behavior: the nature of the reinforcement that establishes and maintains it.
Skinner defined verbal behavior as behavior that is reinforced through the mediation of
another person. He further qualified this by stating that the individual whose action mediates
the reinforcement must have been specifically trained to provide such reinforcement.
Therefore, in order to master the concept of verbal behavior, the concept of mediated
reinforcement must be learned first.
1-1 MEDIATED REINFORCEMENT
Definition: MEDIATED REINFORCEMENT is a stimulus change that has the following
features:
DEFINING
1. It occurs after a response has been emitted.
2. It increases the future probability of that response.
3. It results from the action of another individual.
IRRELEVANT
1. Whether or not the stimulus is conditioned or unconditioned
4
2. The type of receptor affected by the stimulus change
3. Any evocative or eliciting effects of the stimulus change
An example of mediated reinforcement is when you emit the vocal response "water"
and someone then presents you with a glass of water, and this tends to increase or maintain
your tendency to say "water" whenever water would currently be reinforcing. The key feature
is that the reinforcement was provided by the actions of another person. The drink of water
that reinforces turning on the faucet and filling a glass is not an example of mediated
reinforcement, because the reinforcement was not provided by another individual. If water is
presented first, and then the word "water" is emitted, the water is not mediated
reinforcement, because it lacks feature number one of the definition. In short, it is not
reinforcement at all, because the water doesn't come after the response "water.” If the
presentation of the water results in either no increase or a decrease in the tendency to say
"water,” this lacks feature number two and is not an example of any kind of reinforcement.
Another example of mediated reinforcement is when receiving a token from somebody after
emitting the vocal response "token" increases the frequency of the response “token.”
The reinforcement need not be named by the verbal response. If a young child says
"dog" in the presence of the family dog, a parent might say, "Yes, that’s a dog.” This is
mediated reinforcement if the parent's affirmation increases the future probability of this
response under similar circumstances.
1-1 STUDY FRAMES
Write MEDIATED REINFORCEMENT, NONMEDIATED, or NEITHER in the blank after each
example below. Then record your answers by doing the activity for this Study Frame Set on
the CD.
1. You emit the response "milk" and someone presents you with a glass of milk. This
increases the probability that you will say "milk" in the future under similar circumstances.
___________________________
2. You write the word water as a result of hearing someone say "water" and someone says
"correct.” This increases your tendency to write water as a result of hearing someone say
"water." ___________________________
3. Someone presents you with a picture of a dog, and you say "dog.” Presenting the picture
of the dog is an example of ___________________________ .
4. You write the word milk as a result of hearing someone say "water" and someone says,
"That's dumb!" The effect is a decrease in your tendency to write milk as a result of hearing
5
someone say "water.” Someone saying "That's dumb" is an example of
___________________________
5. You turn on a light switch and the lights come on. This results in an increase in your
tendency to turn on lights by turning on the switch. The lights coming on is an example of
___________________________ .
6. A boy says, “sit” and the dog in front of him sits. This increases the probability that the
boy will say “sit” in similar situations in the future. ___________________________
7. A teenager learning to drive shifts smoothly into second gear. Her father says, “Great, I
think you’re getting the hang of this!” This increases her tendency to shift smoothly in the
future. ___________________________
8. A driver asks his passenger to turn up the volume on her FM receiver and the passenger
complies. This increases the probability that the driver makes similar requests in the future.
___________________________
9. A man says “Hello, Clara.” Clara then says, “Aren’t you looking handsome today?” The
man never again greets Clara when he sees her. ___________________________.
10. You gather flowers and place them in a vase in your room. The smell of the flowers fills
the room. This increases the probability that you gather flowers in the future.
___________________________
6
1-2 VERBAL BEHAVIOR
Definition: VERBAL BEHAVIOR is behavior that has the following features:
DEFINING
1. It was established and is maintained by reinforcement.
2. The reinforcement is mediated by another individual.
3. The other individual’s action must have been specifically trained to provide
reinforcement for such behavior.
IRRELEVANT
1. The topography of the behavior: which muscles are used in making the response
2. Dynamic characteristics of the response: speed, intensity, repetition
3. Verbal or nonverbal antecedent stimulus
4. Reinforcement features: conditioned, unconditioned, type of schedule
The first example of mediated reinforcement in the previous section is an example of
the reinforcement of verbal behavior. Saying "milk" and then receiving some milk as the result
of the actions of another person is an example of verbal behavior, because the individual who
provides the milk when hearing “milk” must be trained to do that. In other words, the
individual who mediates the reinforcement for verbal behavior has to have been trained to
respond as a listener when he hears “milk.” The milk that follows the speaker’s response of
saying “milk” maintains that response.
Here are some examples of behavior that is not verbal behavior. Salivation at the sight
of a steak is not verbal behavior, because the salivating was not established and maintained by
reinforcement. Nor is getting a glass of water for yourself verbal behavior, because the
reinforcement is not mediated by the action of another individual. Squirting lemon juice into
someone's mouth is not verbal behavior either, even if the resulting salivation reinforces the
squirting behavior. This is because the listener's action (salivation) was not specifically trained
in order to reinforce the squirting.
Since the type of topography is irrelevant to the definition of verbal behavior, we could
also write the word water, or use sign language, or tap out the Morse code for w..a..t..e..r.
Whether we shout or whisper makes no difference. Whether the form of the response is
controlled by a state of deprivation (a long time without water) or by a prior stimulus (such as
7
the written word water) also makes no difference. Neither does the schedule of
reinforcement make a difference. The verbal behavior may be reinforced every time it occurs,
or only occasionally after some period of time or after many occurrences.
1-2 STUDY FRAMES
Answer by writing either VERBAL BEHAVIOR or NONVERBAL BEHAVIOR in the blank. Then
record your answers by doing the activity for this Study Frame Set on the CD.
1. You enter a very cold room, see that the window is open, and say, "Close the window,
please." Someone else in the room then gets up and closes the window. This results in an
increase in your tendency to say, "Close the window" in the future under similar conditions.
This is an example of ______________________________________.
2. Someone asks you how many cookies you would like and you hold up two fingers. In the
past when you did this, you got two cookies. Holding up two fingers is an example of
____________________________________ .
3. You are walking on some ice and you slip and fall. Falling is an example of
_______________________________________ .
4. You ask someone for a cigarette and he refuses to give it to you. In the past, you
sometimes have received a cigarette when you asked. Asking for it now is an example of
______________________________________.
5. As a result of the person refusing to give you a cigarette, you go out and buy a pack, open
it, and take out a cigarette. Taking out the cigarette is an example of
_________________________________________ .
6. Your ophthalmologist shines a bright light in your eye to see if your pupil contracts.
Shining the bright light is an example of ________________________________
7. You look up a phone number in the directory and then copy it down. Copying the phone
number is an example of ____________________________________.
8. You tie a neon pink ribbon to a tree while hiking. Someone from the camping store told
you this would allow you to make your way back to camp easily. Tying the ribbon to the tree is
___________________.
9. A cat meows and its owner then puts food in the cat’s dish. This increases the future
probability that the cat meows when its owner is present. The cat’s meowing is an example of
__________________________.
10. A developmentally disabled boy who is thirsty points to a picture of a glass of juice and his
mother brings him a glass of juice. The boy is more likely to point to that picture when thirsty.
The boy’s pointing to the picture is ___________________________.
8
TYPES OF CONTROLLING VARIABLES
Before beginning the classification of elementary verbal relations, students must master
several concepts that are used in the definitions of those relations. The concepts concern the
classification of stimuli and responses as either verbal or nonverbal and also the classification
of several types of environmental events that enter into verbal relations.
In the previous section, you learned that in order to identify behavior as verbal
behavior, the behavior must be learned as a result of reinforcement, the reinforcement that
accounts for the acquisition of the behavior must be mediated by another individual (a
listener), and the listener must have been trained specifically to reinforce such activity. In the
chapters that follow, you will learn the names of seven types of elementary verbal relations.
Each type is defined in terms of a relation between a particular kind of antecedent controlling
variable and verbal responses controlled by that type of controlling variable.
There are three general types of antecedent variables that can come to control
occurrences of verbal behavior. Most verbal behavior is controlled by a prior discriminative
stimulus. The discriminative stimuli that control verbal instances may be verbal stimuli or they
may be nonverbal stimuli. Several kinds of controlling relations between discriminative stimuli
and response classes will be discussed in the next chapter.
Another type of controlling variable is called an establishing operation. An establishing
operation is an environmental event or operation that momentarily increases the reinforcing
effectiveness of a particular kind of stimulus change that has reinforced a particular response
class in the past. When an establishing operation is present, it controls occurrences of
behavior that have resulted in that kind of stimulus change in the past. The establishing
operation will be presented in greater detail later.
The third type of antecedent variable that comes to control occurrences of verbal
behavior is called an audience. The presence of any listeners at all, or of particular listeners, or
certain other kinds of stimuli, can determine whether any verbal behavior at all, or which
verbal behavior, occurs at any particular time. The role of an audience in the analysis of verbal
behavior will be discussed in a later section.
In later chapters, we will introduce the seven elementary verbal relations and
distinguish them in terms of the types of antecedent variables that control responses in the
different kinds of relations. In the remainder of this chapter, we will discuss some of the
concepts that you will need to know in order to understand verbal behavior as a special kind of
social interchange in a verbal community composed of individuals who function in the
community both as “speakers” and as “listeners.”
1-3 VERBAL STIMULI
9
Definition: A VERBAL STIMULUS is a physical energy change capable of affecting an
organism's sensory receptors, with the following features:
DEFINING
1. It has a specific form or pattern that as a unit has controlling effectiveness.
2. It is a product of verbal behavior.
IRRELEVANT
1. The modality of the stimulus (the receptors affected by it): auditory, visual, etc.
2. Dynamic characteristics of the stimulus, such as size, intensity, etc.
3. The function of the stimulus: can be discriminative, reinforcing, punishing, etc.
When you say "dog," you produce sound waves that result in an auditory stimulus that
can be heard by a listener -- either the speaker (yourself) or someone else. The auditory
stimulus "dog" is a verbal stimulus, because it has a particular form (the “d” sound first, the
“aw” sound next, and then the “d” sound), and the sound is the result of the action of a
speaker (that action includes pushing air from the diaphragm while moving the lips and tongue
in a very precise way). A photograph of a dog, or an actual dog itself, is a nonverbal stimulus,
because it was not produced by a verbal response.
A verbal stimulus may have features that are nonverbal. For example, if you say “dog”
very loudly and someone says "too loud,” the response "too loud" is not controlled by the
form of the auditory stimulus produced by your verbal behavior, but rather by a dynamic
characteristic of that stimulus (the volume of the sound). The response modality of the verbal
behavior that produces a verbal stimulus is not important. Our first example involved vocal
verbal behavior, but if we had written the word rather than saying it, then the resulting visual
pattern dog could be a verbal stimulus.
1-3 STUDY FRAMES
Write either VS (for verbal stimulus) or NVS (for nonverbal stimulus) in the blank Then record
your answers by doing the activity for this Study Frame Set on the CD.
1. Someone picks up a rock and throws it at you. The sight of the throwing motion and the
approaching rock can be considered a visual stimulus that is _________ .
10
2. The same person throws a rock at you, but you don't see him. Your friend says "duck."
The auditory stimulus that results from your friend's vocal behavior is a _________ .
3. You are driving down the road and see a sign that says, "Beware of falling rocks." The
sign is a visual stimulus that is a _________ .
4. A bee when it flies produces an auditory stimulus that we call a buzzing sound. The
buzzing sound is a _________.
5. A doughnut looks a lot like the letter O and you tend to say "o" whenever you see a
doughnut. The doughnut is a _________ .
6. You see Pepsi in skywriting and say "Those are really large letters." The size of the letters
is a visual stimulus feature that is a _________.
7. You see Pepsi in skywriting and say “Pepsi.” The stimulus that evokes the response
"Pepsi" is a _________.
A grandfather clock chimes 6 times at 6:00 a.m. The tones are a _________.
The washer in the next room starts making a banging sound, because it was loaded unevenly.
After you hear the sound, you go and balance the washer load. The sound of the washer is a
_________.
While you and your friend are looking at a picture of Venus, your friend says, “Mars.” You then
say, “No, that’s Venus.” Your friend saying “Mars” is a _________.
11
1-4 ESTABLISHING OPERATION
Definition: An ESTABLISHING OPERATION is an environmental change or event that has
the following features:
DEFINING
1. It precedes the response it controls.
2. It increases the effectiveness of a particular stimulus change as reinforcement.
IRRELEVANT
1. The type of environmental event
2. The cause of the environmental event
3. The type of stimulus that gains reinforcing effectiveness
Perhaps the most common types of establishing operations, those mentioned by
Skinner in Verbal Behavior, are deprivation and antecedent aversive stimulation. In a situation
where you have had no water for an extended period of time, any response that results in
water is likely to increase in frequency. In other words, the appearance of water is a stimulus
change that reinforces the behavior that produced that change. The appearance of water may
be brought about by many different kinds of behavior, including saying “water, please.” The
response of saying “water!” is verbal behavior, because that response form was acquired as a
result of mediated reinforcement. However, this stimulus change is likely to function as
reinforcement only if you are currently water deprived. Therefore, it is possible to consider
the controlling variables for the response "Water, please" as involving two factors: a necessary
level of deprivation, and discriminative stimuli, typically the presence of a listener who is in a
position to act in a way that gets water to you. Saying “water” may occur in the absence of any
listener if the level of deprivation is sufficient, as in the case of a man dying from thirst in the
desert.
Deprivation, as an establishing operation, is typically related to substances that are
biologically required by an organism, such as food, water, air, heat, and possibly sexual
contact. We can be deprived of many other things without a resultant increase in the
reinforcing effectiveness of a related stimulus change. If we have not had a pencil for a long
period of time, we are not “pencil deprived” and we don’t become increasingly likely to say
“pencil.”
Another class of events that often function as establishing operations is aversive
stimulation. If someone shines a bright light in our face, the removal of the light is likely to
function as a reinforcer, and any behavior that removes the light is reinforced by that stimulus
change. If our muscles ache, the lessening of the aching is likely at that time to have
12
reinforcing function, and any behavior that reduces the aching is reinforced by that stimulus
change.
Another general class of events that function as establishing operations includes those
events that make it temporarily necessary to obtain an item that will allow a reinforceable
response to be made. If someone says that he will give you ten dollars for a recognizable
sketch of a cat, the likelihood that you will ask for a pencil is sharply increased. The
reinforcement value of a pencil has suddenly been increased by the offer. In order to get the
ten dollars, you must sketch a cat, but you can’t do that without a pencil or some other object
necessary for sketching.
The concept of establishing operation is difficult to understand, because it is easily
confused with a discriminative stimulus. Both have the effect of making some particular
behavior likely to occur when they are present. But they work in different ways. Imagine a
water vending machine that costs nothing to use, but only works when the word water is lit
up, which occurs at various times during the day. Assume you have had no water for twenty-
four hours. The light is off and you wait quietly without pushing the button. As soon as the
light comes on, you push the button, and you get a one-ounce cup of water. You push the
button again and get another ounce of water, but then the light goes off, so you wait until it
comes on again before pushing the button again. The light is a discriminative stimulus for
pushing the button. You could push the button anytime, but water follows button pushing
only when the light is on. So you don’t push while the light is off and you do push when the
light is on. After you have had several ounces of water, you don’t push the button any more,
even though the light remains on. That’s because you are no longer water deprived. The
establishing operation is no longer in effect. After awhile you eat a bag of salty peanuts. The
light is still on, so you push the button. Why did you do it only after you ate the peanuts? The
salty peanuts established water as something worth producing, so the behavior that has
produced water occurs. But it only occurs if the light is on.
The light is a discriminative stimulus, because water is available only when it is on. The
water deprivation and the salty peanuts are establishing operations, because water is valuable
only after eating salty food or going awhile without water. If the water is available (light is on)
but is not currently valuable, the behavior doesn’t occur. If the water is not available (light is
off), the behavior doesn’t occur even if water is very valuable. Only if water is valuable and
available does button pressing occur. Control over behavior by these two kinds of antecedents
allows behavior to be very efficient.
1-4 STUDY FRAMES
Write either EO (ESTABLISHING OPERATION) or SD (DISCRIMINATIVE STIMULUS) in the blank.
Then record your answers by doing the activity for this Study Frame Set on the CD.
13
1. You get out a cigarette and then find that you have no matches. Having a cigarette that
needs to be lit is an example of ___________.
2. In order to complete your assignment, you must write examples of establishing
operations and are having a difficult time making an example. The unfinished assignment is an
example of ___________.
3. You notice that a grass fire is burning in the neighbor’s yard. The grass fire functions as
___________ for saying “water.”
As the result of seeing someone place a glass of water on the table in front of you, you say
"water” (perhaps as part of a longer utterance) ___________.
You are stopped at a red traffic light and the light turns green. You put your foot on the gas.
The car moves forward. The light turning green is an example of a(n) ___________.
You are at a friend’s house and note pressure on your bladder. You say, “May I use your
bathroom please?” The pressure on your bladder is an example of a(n) ___________.
You receive a message on your answering machine that says your photos are ready. You then
go to the photo shop. The message is an example of a(n) ___________.
At the holiday party, you see a present with your name on it. You begin to open it. Your name
being written on the present is a(n) ___________ for opening it.
You walk into your house and find that the temperature is 42o F. You turn on the heater. The
room temperature in this example is a(n) ___________.
You are in a lit room. Your friend starts a slide show. You then turn off the light in the room.
The start of the slide show is an example of a(n) ___________.
After tasting your soup, you say to the waiter, "May I have some salt, please?" The waiter is an
example of a(n) ___________.
After tasting your soup, you say to the waiter, "May I have some salt, please?" The unsalted
soup is an example of a(n) ___________.
14
1-5 POINT-TO-POINT CORRESPONDENCE
Definition: POINT-TO-POINT CORRESPONDENCE is a relationship between a verbal
discriminative stimulus and the response it controls with the following features:
DEFINING
1. The discriminative stimulus must have two or more components.
2. The response must have two or more components.
3. The first component of the stimulus must control the first part of the response; the
second part of the stimulus must control the second part of the response, etc.
IRRELEVANT
1. Formal or dynamic characteristics of the stimulus
2. Formal or dynamic characteristics of the response
The vocal response "cat" made as a result of hearing someone say "cat" has point-to-
point correspondence between the stimulus and the response. The "k" sound controls the first
part of the response; the "a" sound controls the second part of the response; and the "t"
sound controls the final part of the response. If you said "cat" as the result of hearing
someone say "feline,” there would be no point-to-point correspondence between the stimulus
and the response. "Feline" has five sounds or components while "cat" only has three. Making
the sound "b" as a result of hearing someone say "b" also is not an example of point-to-point
correspondence, because neither the stimulus nor the response has two or more
components. Seeing the letter b and saying "bee" also is not an example of point-to-point
correspondence, because the stimulus only has one component. Saying "dog" as the result of
seeing the word dog is an example of point-to-point correspondence. It makes no difference
whether the response is vocal or written or whether the stimulus is auditory or
visual. Writing dog as the result of hearing someone say "dog" would illustrate point-to-point
correspondence. Dynamic features are also irrelevant. If you shouted "DOG" as the result of
hearing someone whisper "dog" that would still be point-to-point correspondence.
The correspondence is not necessarily between letters. If you see the word write and
say "write," there are probably only two components involved. The visual wr controls the
vocal "r" sound and the visual itecontrols the vocal "ite.” This is one reason it is hard to learn
to read words like write and right. In copying the word write, each letter would be a
component.
1-5 STUDY FRAMES
15
Write PTP (POINT-TO-POINT) or NONE in each blank. Then record your answers by doing the
activity for this Study Frame Set on the CD.
1. The vocal response "cat" is controlled by the written word cat. __________
2. The vocal response "dog" is controlled by the written word cat. __________
3. The vocal response "cat" is controlled by the auditory stimulus "cat.” __________
4. The stimulus is the written letter a and the response is the written response A.
__________
5. The stimulus is the written letter w and the response is the vocal response “double-you.”
__________
6. You write the word cite as a result of seeing the written word sight. __________
7. The vocal response “fort” is controlled by the written word FORT. __________
8. The written response FORT is controlled by the auditory stimulus “fort.” __________
9. The written response FORT is controlled by the written response BASE. __________
10. The vocal response “eff-oh-are-tee” is controlled by the auditory stimulus “fort.”
__________
16
1-6 RESPONSE-PRODUCT
Definition: A RESPONSE-PRODUCT is a stimulus with the following features:
DEFINING FEATURES:
It is the result of an organism's behavior (verbal or non-verbal).
IRRELEVANT FEATURES:
Formal or dynamic characteristics of the stimulus (e.g., mode or intensity)
Formal or dynamic characteristics of the response that produces it
The function of the stimulus (e.g., discriminative, reinforcing, etc.)
Virtually all behavior produces a change in the environment that can have a stimulus function.
Any movement that an organism makes results in a change in the visual stimulation affecting
another organism who is observing the action. Many actions also result in the production of
other stimuli. Certain movements of the vocal musculature result in auditory stimuli (sounds).
Behavior that results in contact with another organism may result in stimulation of the other
organism's tactile receptors (for example, the skin). Behavior may result in permanent or
relatively permanent stimuli. For example, the hand and arm movements involved in writing
with a pen usually leave visible marks on a surface. A response may have more than one
product. The same writing that produces marks on a piece of paper also produces visual stimuli
related to arm movements; we can see that someone is writing even when we can't see the
marks on the paper. Furthermore, if it is quiet, some auditory stimuli may be produced as a
result of the friction between the pen and the paper. Additional response-products may be
private; the movements resulting from writing also produce kinesthetic stimulation that the
writer may be able to react to.
Of course, not all forms of stimulation result from someone's behavior. The physical
environment also changes and these changes result in the production of stimuli. Rain falling
from the sky results in visual stimulation, auditory stimulation when it strikes the ground or
windows, and tactile stimulation if it strikes someone's skin. If the rain is cold, it may also
affect thermoreceptors on the surface of the skin. Also, we are talking about stimuli that are
the direct and immediate product of a prior response. If someone starts a motion picture
projector, the visual stimuli that are the result of the scenes changing on the screen are not to
be considered response-products of the behavior of turning on the projector. The response-
products of turning on a projector include the visual stimuli that result from the movement of
the wrist and hand, auditory stimuli produced if the machine clicks when it is turned to the
"on" position, and even visual stimuli that result from the change in position of the "on-off"
switch or knob.
1-6 STUDY FRAMES
Write either RPV (RESPONSE-PRODUCT OF VERBAL BEHAVIOR), RPNV (RESPONSE-PRODUCT OF
NON-VERBAL BEHAVIOR),or NRP (NOT A RESPONSE-PRODUCT) in the blank. Then record your
answers by doing the activity for this Study Frame Set on the CD.
17
1. The auditory stimuli produced by someone who is standing behind you saying "dog."
__________
2. The visual stimulation produced by a deaf person finger signing d-o-g. __________
3. The visual stimulation produced by the wagging of a dog's tail. __________
4. Auditory stimulation that results from windshield wipers moving back and forth that
sounds like the phrase "You are late." __________
5. The auditory stimulation produced by hail that breaks your window. __________
6. The auditory stimulation produced by a child's squealing on a roller coaster. __________
7. The visual stimulation produced by the sun disappearing below the horizon . __________
8. The visual stimulation produced by a traffic cop's arm signals. __________
9. The tactile stimulation produced by someone tapping Morse code on your hand.
__________
10. The tactile stimulation produced by a vibrating pager. __________
18
1-7 FORMAL SIMILARITY
Definition: FORMAL SIMILARITY is sameness in the formal properties of a response
product and the stimulus that preceded that response:
DEFINING
1. The stimulus and the response-product are both in the same modality (e.g., they are both
visual or both auditory).
2. Their physical patterns or sequences resemble one another (e.g., they look or sound
alike).
IRRELEVANT
1. Specific formal or dynamic features of both the stimulus and the response-product
2. The number of additional formal or dynamic features resembling one another
Someone says "horse" and as a result of that auditory stimulus, you also say "horse.
Formal similarity exists in this case between the auditory stimulus "horse" and the response-
product of the vocal response "horse.” The pattern of auditory stimuli is the same, with the
"h" sound at the beginning and the "s" sound at the end. Some degree of formal similarity
exists even if other formal or dynamic features are somewhat different. For example, the
stimulus may have been low-pitched, whereas, the response-product may have been very
high-pitched. Another example might involve a stimulus that was produced very rapidly, and
the response-product may be the result of a response made somewhat more slowly. If in fact
the pitch is about the same and the speed is about the same, then it could be said that a
greater degree of formal similarity exists than when the pitch or speed is different.
The existence of formal similarity, even minimally, allows for a unique form of self-
reinforcement, which makes the acquisition and maintenance of responses that produce
formal similarity between their products and their controlling stimuli more likely to occur. This
essentially allows for a form of self-correction. In the case of verbal behavior, the two general
categories in which this can occur are when we repeat what has just been said, or when we
copy what someone has written. In either case, we can correct our response if the similarity is
not adequate. This same type of self-correction is possible in any type of vocal mimicry and/or
type of copying.
Formal similarity may exist when the stimulus and response-product have only one
component. When formal similarity exists between a stimulus and a response-product where
both have more than one component, point-to-point correspondence must exist between the
stimulus and the response it controls.
19
1-7 STUDY FRAMES
Write FS (FORMAL SIMILARITY), PTP (POINT-TO-POINT CORRESPONDENCE),
B (BOTH), or N (NEITHER) in the blank. Then record your answers by doing the activity for this
Study Frame Set on the CD.
1. Saying "response" as a result of hearing the word "stimulus." __________
2. Saying "stimulus" as a result of hearing the word "stimulus." __________
3. Saying "stimulus" as a result of seeing the written word stimulus. __________
4. Writing the word stimulus as a result of hearing it spoken. __________
5. Writing the word stimulus as a result of seeing the written word stimulus. __________
6. Writing the number 8 as a result of seeing the number 8 written on a blackboard.
__________
7. Signing (finger-spelling) the word “daughter” as a result of seeing the written
word daughter. __________
8. Saying the word “daughter” as a result of hearing the Morse code for the word
“daughter.” __________
9. Writing the word daughter as a result of seeing it written on a blackboard. __________
10. Signing the letter “D” as a result of seeing it signed. __________
20
1-8 FORMAL AND THEMATIC CONTROL
Definition: FORMAL CONTROL exists whenever a controlling variable evokes a response
and the relation has the following features:
DEFINING
1. There is point-to-point correspondence or formal similarity between the controlling
variable and the response.
IRRELEVANT
1. Formal characteristics of either the controlling variable or the response
2. Dynamic characteristics of either the controlling variable or the response
Definition: THEMATIC CONTROL describes a situation in which a controlling variable
evokes a response and the relation has the following features:
DEFINING
1. There is NO point-to-point correspondence OR formal similarity between the controlling
variable and the response.
IRRELEVANT
1. The same as the irrelevant features for formal control.
A tendency to say what you have just heard illustrates formal control. Someone says
"Say alligator" and you then say "alligator.” Notice that everything you heard did not have to
be repeated for formal control to exist between your response and one or more of the
controlling stimuli. If you were to respond to the same request by saying "crocodile," this
would then be an example of thematic control, because there would be no point-to-point
correspondence or formal control between the stimulus and the response. For that matter,
you could say "door knob" and that also would be thematic control. If you say "alligator" as a
result of seeing the written word alligator, this is also an example of formal control even
though there is no formal similarity, because there is point-to-point correspondence between
the stimulus and the response. If you see an alligator and then say "alligator,” the type of
control is thematic, because there is no point-to-point correspondence between the stimulus
and the response. The head of the alligator does not control the "al" and the tail does not
control the "gator.” If you write the word alligator when you see an alligator, the control is still
thematic, even though the stimulus and the response-product are both visual stimuli. The
21
concepts of formal and thematic control play an important role in several of the aspects of the
analysis that follows.
1-8 STUDY FRAMES
Write either FORMAL or THEMATIC in the blank following each of the examples. Then record
your answers by doing the activity for this Study Frame Set on the CD.
1. Saying "response" as a result of seeing the word stimulus. _______________
2. Saying the word "stimulus" as a result of hearing the word "stimulus.” _______________
3. Writing stimulus as a result of seeing the written word stimulus. _______________
4. Saying "water" as a result of water deprivation. _______________
5. Writing water as a result of seeing a glass of water. _______________
6. Saying “Kansas” as a result of seeing the word Kansas on a map. _______________
7. Saying the word “Kansas” as a result of seeing Kansas on a map. _______________
8. Saying the word “Kansas” as a result of hearing the word “computer.” _______________
9. Writing the word “Kansas” as a result of seeing the word “Wichita.” _______________
10. Writing the word “Kansas” as a result of hearing the word “Kansas” _______________
22
CHAPTER 2: ELEMENTARY VERBAL RELATIONS
Introduction
There are several types of elementary verbal relations. They are called “elementary”
because verbal relations of these kinds are the “building blocks” for more complex relations
between the environment and verbal responses. They are also the building blocks that make
possible novel verbal response forms and novel controlling relations between the environment
and earlier acquired response forms. In all cases, the elementary verbal relations that come to
exist in a human’s verbal repertoire are the result of a prior history of reinforcement. If they
weren’t, the relations would not be designated as verbal for purposes of this analysis.
Each type of verbal relation gets its name from the relation between a type of prior controlling
(antecedent) variable and a subsequent response. As you learned at the end of the last
chapter, there are two general types of verbal controlling
relations: formal and thematic. Formal control exists when there is point-to-point
correspondence between a prior stimulus and a verbal response. For all practical purposes
this means that formal control always involves a relation between an antecedent verbal
stimulus and a verbal response. Thematic control exists when there is no point-to-point
correspondence between a verbal response and a prior controlling event. The prior controlling
event may verbal or nonverbal.
In this chapter, we will discuss first four types of elementary controlling relations that
involve formal control. Then we will discuss three types of elementary controlling relations
that involve thematic control. We will also discuss a fourth type of thematic control,
called audience control, which is more general. The seven kinds of elementary controlling
relations are seen in a great deal of verbal behavior that can be casually observed in the
natural environment. They are the foundation for complex activities such as writing papers
and books, making speeches, teaching, interpersonal communication, etc. Audience control is
almost always a relevant variable when verbal behavior occurs, because verbal behavior is not
reinforced in the absence of an audience.
23
VERBAL RELATIONS: FORMAL CONTROL
2-1 ECHOIC VERBAL RELATION
Definition: An ECHOIC is a verbal relation that has the following features:
DEFINING
1. The response is vocal (response product is auditory).
2. The response is controlled by a prior auditory verbal stimulus.
3. There is formal similarity between the stimulus and the response-product.
4. There is point-to-point correspondence if both stimulus and response have two or
more components.
IRRELEVANT
1. The particular form of the stimulus and the response-product
2. Dynamic characteristics of both the stimulus and the response
3. The meaning of either the stimulus or the response
4. Whether or not the current response is reinforced
In everyday terms, echoic behavior is saying what was just heard (“echoing”). Saying “echoic”
upon hearing someone say “echoic” is an example of echoic behavior. So is saying “oh-wah-ta-
goo” upon hearing “oh-wah-ta-goo.” Saying “echoic” as a result of seeing the word is not
echoic; nor is saying it as a result of hearing someone say “imitative.” Sometimes there is
more formal similarity between the stimulus and the response-product than is required to
meet the definition of echoic. For example, the skilled mimic not only duplicates the speech
sounds but uses the same intonation, pitch and volume as the original speaker.
The dynamic features of both the stimulus and the response are irrelevant. The stimulus may
be very loud and the response very soft. Whether or not the response has any meaning does
not enter into the definition. Whether or not the current response is followed by a reinforcer
is not relevant to identifying a verbal relation as echoic.
24
2-1 STUDY FRAMES
Write either E (ECHOIC) or NON (NON-ECHOIC) in the blank. Then record your answers by
doing the activity for this Study Frame Set on the CD.
1. Saying “ice cream” as a result of hearing someone say “ice cream.” __________
2. Saying “dog” as a result of seeing a dog. __________
3. Someone tells us his name and we repeat it so that we will remember it.
__________
4. Your instructor writes the next assignment on the blackboard and then you write
it in your notebook. __________
5. Saying “stop” as a result of seeing a stop sign. __________
6. You see someone yawn and as a result find yourself yawning. __________
7. Saying “di-oxy-ribo-nu-cleic” slowly and carefully after hearing the instructor say
“dioxyribonecleic acid” as if it rolled off his tongue. __________
8. Writing the word ice cream as a result of hearing someone say it. __________
9. Saying “ahhh” when the physician says “Say ‘ahhh’.” __________
10. You say “ice cream” as the result of wanting some. __________
11. You say “ice cream” as the result of hearing someone say “popsicle.” __________
12. Saying “ZEB” as the result of hearing someone say “ZEB.” __________
25
2-2 COPYING A TEXT
Definition: COPYING A TEXT is a verbal relation with the following features:
DEFINING
1. The response is writing, printing or typing and the response-product is visual.
2. The controlling variable is a product of someone’s prior verbal behavior of writing,
typing or printing.
3. There is point-to-point correspondence between a multi-component controlling
variable and the response.
4. There is formal similarity between the controlling variable and the product of the
response (in some cases there is little or no formal similarity, but these can be considered
exceptions to the general rule).
IRRELEVANT
1. Formal and dynamic features of the response
2. Formal and dynamic features of the controlling stimulus
3. Whether or not the current response is reinforced
4. The “meaning” of either the stimulus or the response
Anytime you copy something that someone, including yourself, has written, that is
called copying a text. If you see Test next Friday written on the blackboard and you then write
the same thing in your notebook, that is an example of copying a text. It does not matter
whether the message was printed on the blackboard and you wrote your notes in cursive, as
long as there is point-to-point correspondence between the stimulus and the
response. Writing the word table as a result of seeing la mesa written is not copying a
text. Although the stimulus and the response are in the right modes, there is no point-to-point
correspondence between prior stimulus and response. Writing table as a result of hearing
“table” is also not copying a text, but is a type of elementary verbal relation that will be
presented next. It doesn’t matter whether the letters in the stimulus are huge and the writing
responses are tiny or vice versa. It also doesn’t matter whether the stimulus is written,
printed, or typed, because the response-products of a written word have approximately the
same formal characteristics as a typed word. Similarly, it doesn’t matter whether the response
is printed, typed, or tapped out in Morse code, etc.
Sometimes only part of a longer text is copied, but copying that part is copying a text if the
relevant features are present. For example, you may have written Test Friday, or
just Test! after seeing Test Next Friday on the board. Copying test and copying Friday both
have all the defining features of copying a text.
2-2 STUDY FRAMES
26
Write E (ECHOIC), (CT) COPYING A TEXT, or N (NEITHER) in the blank. Then record your
answers by doing the activity for this Study Frame Set on the CD.
1. Saying “chili” as a result of seeing the word chili written. __________
2. Writing chili as a result of hearing the word “chili.” __________
3. Salivating as a result of hearing the word “chili.” __________
4. Writing the word chili as a result of seeing the typed word chili. __________
5. Writing chili as a result of seeing the word enchilada. __________
6. Saying the word “chili” as a result of hearing someone say “let’s have chili.”
__________
7. Printing the word CHILI in capital letters as a result of seeing the word chili written
in script. __________
8. Writing the words Dr. Pepper as a result of seeing them written with ten-foot-high
letters on a billboard that reads Drink Dr. Pepper at 10, 2 and 4. __________
9. Writing BEZ as the result of seeing BEZ written on a blackboard. __________
10. Saying “Dr. Pepper” as a result of hearing someone say, “Drink Dr. Pepper.”
__________
27
2-3 TAKING DICTATION
Definition: TAKING DICTATION is a verbal relation that has the following features:
DEFINING
1. The modality of the response is writing (printing, typing, etc.).
2. The controlling variable is the response-product of someone’s prior vocal (or
finger-spelled) verbal behavior.
3. There is point-to-point correspondence between the stimulus and the response.
IRRELEVANT
1. Formal and dynamic characteristics of the stimulus
2. Formal and dynamic characteristics of the response
3. Whether or not the current response is reinforced
4. The “meaning” of either the stimulus or the response
Someone tells you the name of a good restaurant and you write it down; that is taking
dictation. In Skinner’s analysis, taking shorthand is not called taking dictation, because it lacks
the necessary point-to-point correspondence. Writing the name of the restaurant as the result
of seeing the name in the phone book has already been classified as copying a text. It is not
taking dictation, because the controlling variable is not the result of someone’s
prior vocal verbal behavior. Writing diner as the result of hearing someone say “restaurant” is
also not taking dictation, because there is no point-to-point correspondence between the
stimulus and the response. What is said, even if it is nonsense, makes no difference to the
classification. It also doesn’t matter whether the stimulus is loud or soft, or the response is fast
or slow. The response can be either handwritten, printed, typed, etc.
28
2-3 STUDY FRAMES
Identify each of these examples of verbal behavior by writing E (ECHOIC), CT (COPYING A
TEXT), TD (TAKING DICTATION,) or NONE (NONE OF THESE) in the blank. Then record your
answers by doing the activity for this Study Frame Set on the CD.
1. Writing textbook as a result of seeing the word textbook. __________
2. Saying “Did you say ‘Enron’?” upon overhearing someone say, “I’m going to work
for Enron.” __________
3. Writing calendar as a result of hearing someone say “calendar.” __________
4. Writing textbook as the result of hearing someone say “classroom.” __________
5. Writing BAJ as the result of hearing someone say “BAJ.” __________
6. Writing BAJ as the result of hearing someone say
“TEK.”______________________.
7. Saying “My name is Marva” when the teacher says “My name is Marva.”
__________
8. Writing www.hotmail.com when you see that web address in a newspaper
advertisement. __________
9. Writing not echoic after hearing someone say, “I don’t know what the controlling
relation is, but it’s definitely not echoic.” __________
10. Saying “www.hotmail.com” as a result of seeing that web address in a newspaper
advertisement. __________
29
2-4 TEXTUAL BEHAVIOR
Definition: TEXTUAL BEHAVIOR is a verbal relation that has the following features:
DEFINING
1. The response is vocal.
2. It is controlled by a prior stimulus that is the response-product of writing behavior.
3. There is point-to-point correspondence between the stimulus and the response.
IRRELEVANT
1. The specific topography and dynamic characteristics of the vocal response
2. The specific form of the visual verbal stimulus (written, printed, or typed)
3. Whether or not there is any reinforcement for the current response
4. The “meaningfulness” of stimulus and the response
If you look at the title of this page and that evokes the vocal response “textual behavior,” all
three parts of the definition are present and your behavior would be an example of textual
behavior. Hearing someone say “textual behavior” and then saying it yourself is not textual
behavior, because the antecedent stimulus is not the response-product of writing
behavior. You should have identified this as an example of echoic behavior. Writingtextual
behavior as a result of looking at the title of this page is a type of verbal relation that has
already been discussed -- copying a text. It is not textual behavior, because the response is not
vocal and textual behavior requires a vocal response according to defining feature number
1. Looking at the title of this page and then saying “reading” is also not textual behavior, even
though the stimulus is written and the response is vocal. The problem is that there is no point-
to-point correspondence. Saying “KAJ” as a result of seeing KAJ is textual behavior, because all
the defining features are present. We have already noted that the “meaningfulness” of the
stimulus or the response is an irrelevant feature.
30
2-4 STUDY FRAMES
Classify the following examples of verbal behavior by writing TB (TEXTUAL BEHAVIOR), CT
(COPYING A TEXT), TD (TAKING DICTATION), E (ECHOIC BEHAVIOR), or NONE (NONE OF
THESE) in the blank. Then record your answers by doing the activity for this Study Frame Set on
the CD.
1. You write no smoking as a result of seeing a NO SMOKING sign. __________
2. You say “no smoking” as a result of hearing someone else say the same thing.
__________
3. Saying “stupid sign” as a result of seeing a NO SMOKING sign. __________
4. Saying “DAK” as a result of seeing the word DAK written. __________
5. Saying “Dr. Pepper” as a result of seeing a Dr. Pepper can. __________
6. Saying “no smoking” as a result of seeing a NO SMOKING sign. __________
7. Overhearing someone at the next table say that finals begin on December 10, you
write finals December 10 on your notepad. __________
8. Saying “french fries” as a result of seeing french fries listed on a menu.
__________
9. Writing the word fries as a result of seeing french fries listed on a menu.
__________
10. Saying “french fries” as a result of seeing some french fries. __________
31
VERBAL RELATIONS: THEMATIC CONTROL
All of the elementary verbal relations we have studied thus far – echoic, copying a text, taking
dictation, and textual behavior – are examples of formal control; that is, they all are functional
relations in which there is point-to-point correspondence between the controlling stimulus
and the evoked response. Not all elementary verbal relations have such point-to-point
correspondence. You previously learned the term thematic for the kind of antecedent control
where there is no point-to-point correspondence between the stimulus and the
response. What follows are types of elementary verbal relations that are examples of
thematic control. This kind of control is called thematic, because in most cases the controlling
variable strengthens several responses at the same time. This fact will become clearer as each
of the elementary relations involving thematic control is presented. In the relations involving
formal control, each controlling variable strengthens only a single response form.
32
2-5 THE INTRAVERBAL RELATION
Definition: An INTRAVERBAL is a verbal relation having the following features:
DEFINING
1. A verbal response
2. The antecedent controlling variable is a verbal stimulus.
3. There is no point-to-point correspondence between the stimulus and the
response.
IRRELEVANT
1. The musculature involved in executing the response
2. The modality of the prior verbal stimulus (usually either visual or auditory)
3. Dynamic features of either the stimulus or the response
4. The “correctness” of the response
Word association tests are a common example of intraverbal behavior. Saying “response” as a
result of the experimenter saying “stimulus” is an example of an intraverbal
relation. Mathematics provides many good examples. Saying “four” as a result of seeing “two
plus two” is interverbal. Saying “response” as the result of hearing “response” has already
been classified as echoic. It is not intraverbal, because there is point-to-point correspondence
between the stimulus and the response, which violates the third defining feature of the
intraverbal relation. Saying “stimulus” as the result of seeing a light come on is also not
intraverbal, because the prior controlling stimulus is nonverbal. The stimulus may be in any
sensory modality (auditory, visual, tactile, etc.) and the response may be in any response
modality (writing, signing, speaking, etc.) Writing responseas the result of either hearing
someone say “stimulus” or watching someone write stimulus is intraverbal. The modality of
the stimulus and of the response is unspecified in the intraverbal. The final issue is
correctness. Saying “five” as the result of seeing or hearing “two plus two equals” is still
intraverbal, because all three of the defining features are present. The “correctness” of the
response is irrelevant.
33
When you are first learning a foreign language, most of your behavior is intraverbal. The
English “good day” evokes the French “bonjour” and vice versa. Some critics have claimed that
the kinds of examples of intraverbal relations you have just read are trivial examples of
“language.” While it is true that many intraverbal relations are trivial in the sense that they
don’t “say much,” a great deal of the intraverbal behavior we acquire is very useful. For
example, definitions, instructions, and word associations usually help us get along in the
everyday world. Intraverbal behavior also plays a significant role in conversation, although an
ongoing conversation would typically involve several different types of elementary verbal
relations, plus some more complex relations beyond the scope of this text.
2-5 STUDY FRAMES
Write in the blank INT (INTRAVERBAL), E (ECHOIC BEHAVIOR), TB (TEXTUAL BEHAVIOR), CT
(COPYING A TEXT), TD (TAKING DICTATION), or NONE (NONE OF THESE) to
identify yourbehavior in the example. Then record your answers by doing the activity for this
Study Frame Set on the CD.
1. You say “water” as the result of water deprivation. ____________
2. You write water as a result of hearing “bread and …” ____________
3. You say “water” as a result of seeing a lake. ____________
4. You say “water” as a result of seeing the word water written. ____________
5. You say “water” as a result of seeing the word ocean. ____________
You write water as result of seeing water finger-spelled. __________
You type water as a result of hearing the auditory stimulus “l’eau.” __________
You write water as a result of seeing a picture of water. __________
9. You say “water” as a result of hearing someone say “water.” __________
10. You say “water” as a result of hearing a baby say “wa-wa.” __________
11. You say “water” as a result of hearing someone say “zoogleeack.” __________
12. You say “water” as a result of hearing the question, “What’s the best way to put out a
grease fire?” __________
13. You write “water” after seeing the word written in a cookbook. __________
14. You pour a glass of water as a result of someone saying, “Water, please.” __________
34
35
2-6 THE TACT RELATION
Definition: A TACT is a verbal relation that has the following features:
DEFINING
1. The response is verbal.
2. The controlling variable is a nonverbal discriminative stimulus, which is an object
or event or a property (or combination of properties) of an object or event.
IRRELEVANT
1. The musculature with which the response is executed (it may be vocal, written,
gestural, etc.)
2. The modality of the stimulus (it may be auditory, visual, gustatory, olfactory, etc.)
3. Whether or not there is any reinforcement for the current response
4. The correctness of the response
All of the elementary verbal relations we have studied up to this point have had some type of
verbal stimulus as the controlling variable. But what about the vast amount of verbal behavior
whose form is controlled by a prior nonverbal stimulus? Whenever we identify some feature
of our physical environment, the description we give is typically controlled by prior nonverbal
stimuli in the form of objects, events, their properties, or spatial and temporal relations
between them. The relation between any of these kinds of nonverbal stimuli and a verbal
response is called a tact.
Saying “dog” as the result of seeing a dog is an example of the tact relation. If you say “dog”
either because you see the word dog or hear it spoken, then the relation is textual or echoic,
respectively. These latter two cases would not be tacts, because the prior controlling stimuli
are not nonverbal stimuli as is required by the tact definition. Likewise, if you say “bring me
the dog,” the response form “dog” is controlled by the fact that being presented with the dog
would currently be reinforcing; that is, the form of the response is controlled by an
establishing operation, not a nonverbal stimulus.
The response modality is not an important consideration in classifying a verbal relation as a
tact. The response may be vocal, written, gestural, or any other activity, as long as the
reinforcement accounting for its existence in the repertoire is mediated and the form of the
response is controlled by a prior nonverbal stimulus. Furthermore, the controlling antecedent
stimulus may be any type of stimulus (as long as it’s nonverbal) and it may be simple or
complex. That is, the controlling stimulus may be a single property of a complex event, such as
a color, or it may be the entire object, such as a dog, or a city, or a picture of the solar
system. When we say that a speaker uses a word to “refer to” an object, we are usually talking
about a verbal relation fitting the definition of a tact.
36
Correctness is a final issue. Saying “cat” when a dog is the discriminative stimulus is a
tact. Our verbal community calls such a relation “incorrect” and we typically do not reinforce
calling a dog “cat.” But if saying "cat" is controlled by the presence of a dog, the relation
between the nonverbal stimulus and the verbal response is a tact. Tacts that do not meet the
standard reinforcing practices of the verbal community represent a type of extension of the
elementary tact relation. These are discussed in the next chapter.
2-6 STUDY FRAMES
In the first blank, write INT (INTRAVERBAL), E (ECHOIC BEHAVIOR), TB (TEXTUAL
BEHAVIOR), CT (COPYING A TEXT), TD (TAKING DICTATION), TACT, or NONE (NONE OF THESE).
In the second blank, indicate whether the type of control is FORMAL or THEMATIC. Then
record your answers by doing the activity for this Study Frame Set on the CD.
1. Saying “duck” as the result of hearing someone say “soup” is called:
a. ____________________ b. ____________________
2. Saying “goose” as a result of seeing a duck.
a. ____________________ b. ____________________
3. Saying “get down” as a result of a flying object heading toward a friend.
a. ____________________ b. ____________________
4. Someone holds up a card with the word duck on it and you say “duck.”
a. ____________________ b. ____________________
5. Saying “dog” as a result of hearing a dog bark.
a. ____________________ b. ____________________
6. Saying “loud” as a result of hearing someone shout “duck.”
a. ____________________ b. ____________________
7. Typing the word duck as a result of seeing duck written on a blackboard.
a. ____________________ b. ___________________
8. Saying “duck” as a result of hearing someone say “Mallard.”
a. ____________________ b. ___________________
9. Shouting “DUCK!” as a result of someone whispering “Mallard.”
a. ____________________ b. ___________________
10. Saying “a duck” as a result of someone saying, “I used to have a pet duck.”
a. ____________________ b. ___________________
37
11. Writing “duck” as a result of someone saying, “D-U-C-K.”
a. ____________________ b. ___________________
12. Saying “duck” when looking at a photograph of a duck.
a. ____________________ b. ___________________
13. Saying “ducks aplenty!” after seeing a sign on a taxidermy shop reading, We’ve got
ducks aplenty!
a. ____________________ b. ___________________
14. Saying “quack, quack” as a result of seeing a duck.
a. ____________________ b. ___________________
38
2-7 THE MAND
Definition: The MAND is a verbal relation that has the following features:
DEFINING
1. The response is verbal.
2. The form of the response is controlled by an establishing operation.
IRRELEVANT
1. The musculature with which the response is executed (may be vocal, written,
gestural, etc.)
2. The type of reinforcement made currently effective by the establishing operation
(may be conditioned or unconditioned, etc.)
3. Whether or not the current response is followed by a reinforcer
All of the types of elementary verbal relations that you have studied up to this point have been
relations between a prior discriminative stimulus, either verbal or nonverbal, and a verbal
response. Specific controlling relations of these types come to exist in a learner’s repertoire as
a result of the mediated reinforcement supplied by the speaker’s verbal community. Listeners
in a verbal community teach new learners to call dogs “dog” and read dog as “dog” and write
“dog” as dog, etc. They do this by making all sorts of reinforcers contingent on emitting
specific verbal responses if and only if they are appropriate in terms of a stimulus that is
present. Most of the reinforcers mediated by the verbal community to teach new learners
tacts, intraverbals, echoics, textuals, etc. are social events, often verbal, such as various forms
of approval, provision of useful information, points or grades, etc. Important to remember is
that the relation between the sight of a dog and saying “dog” may be reinforced in many
different ways. This is the case for all the controlling relations that involve verbal and
nonverbal discriminative stimuli and specific response forms.
There is one type of elementary verbal relation that is different, both in the way the verbal
community reinforces occurrences of that relation and also in the type of antecedent variable
that comes to control response occurrences. This type of elementary verbal relation is called
a mand. Mands differ in two ways from the elementary verbal relations you have already
learned. First, their response form (vocalized sounds, signs, or written marks) is followed by a
very specific kind of reinforcing consequence. By "specific" we mean that the consequence
that typically follows those responses is the same consequence time and time again. For
example, when we ask for “water,” we characteristically receive water. We do not usually
receive milk or cookies. The verbal community rarely thanks us, gives us points, or approves of
asking for water. Rather, the reinforcer mediated by the verbal community is
water. Similarly, saying "stop" generally results in a halt in a listener's action and it usually
does not result in money, or praise, or approval. Responses that are typically followed by a
specific type of consequence are designated as mands. Responses commonly called demands,
39
commands, and reprimands usually fit the definition of a mand and the classification derives
its name from these terms.
A history of specific reinforcement does not account for the occurrence of a particular
response form at some particular time, any more than a history of more general types of
reinforcement (approval, etc.) accounts for the occurrence of a response form at a particular
time. What the mediated reinforcement does is to bring particular response forms under
control of situations present when a response form occurs. In the case of tacts, the various
consequences work to bring a particular response form under control of features of the
nonverbal environment. In the case of textuals, the various consequences work to bring a
particular response form under control of the verbal stimuli that are read. In all the
elementary verbal relations you have studied so far, the event that comes to control the
response form functions as a discriminative stimulus. Only when that stimulus has been
present have certain response forms resulted in the verbal community’s delivery of
reinforcers. So that stimulus acquires the function of evoking a particular response form.
In order to explain individual occurrences of the verbal responses that
produce specific reinforcers, we need to be able to specify some environmental event that
evokes the response having this particular form. The type of antecedent controlling event in a
mand relation was discussed earlier and it was called an establishing operation. Recall that an
establishing operation alters the current reinforcing value of events that can be produced by
behavior. If water deprivation alters the current reinforcing value of water, then behavior that
can produce water is likely to occur as long as the water deprivation is in effect. If pain alters
the current reinforcing value of reduction of pain, then behavior than can produce pain
reduction is likely to occur as long as the pain is present. So a mand is a verbal relation
between a particular establishing operation and a specific form of verbal response.
The verbal response “water” is a mand when it is controlled by a condition that makes water
something worth producing. The verbal response “water” is a tact when it is controlled by an
antecedent nonverbal stimulus (such as a glass of water, a lake, etc.). And the verbal response
“water” is an echoic when it is controlled by the antecedent verbal stimulus “water.”
There may be several different establishing operations that can control a specific response
form. What they must have in common is that they increase the current value of some event
that has, in the past, reinforced responses having that form. If any of them is currently in
effect, then it is likely to control that response. So “water” may be part of a mand relation in
which water deprivation, or dying flowers, or salty food is the antecedent. If any of these
antecedent conditions control the response “water,” then the relation between that
antecedent and that verbal response is a mand. To identify a verbal relation as a mand, the
form of the response in the relation must be controlled by an establishing operation.
Remember that the response modality is not specified in the mand. You
can say “water,” write water, sign water. The response in a mand relation may be in any
response modality as long as the controlling variable is an establishing operation. The
establishing operation in a mand relation may be some type of conditioned reinforcement,
such as money or a specific form of social reinforcement. If the verbal response “Tell me that
you love me” can only be reinforced by a consequence having the form “I love you,” we should
40
consider what establishing operation might be controlling that response. (In many cases, the
establishing operation is likely to be some sign that the listener is inclined to escape or has
turned attention from the speaker, which establishes the affirmation as a current
reinforcer.) Whether the manded verbal reply does or does not follow the mand on this
occasion is an irrelevant feature.
41
2-7 STUDY FRAMES
Write the name of any of the elementary verbal relations that you have learned up to this
point, including the MAND. Some examples may be NONE (NONE OF THESE). Then record your
answers by doing the activity for this Study Frame Set on the CD.
1. You have an important appointment and when you check your watch you find that
it has stopped. You ask a person who is standing nearby, “What time is it?”
________________________
2. You’re watching TV and a horrible singing commercial comes on. You grab the
remote and turn the volume down so that you can no longer hear the commercial. Turning
the volume down is ________________________.
3. A child is acting very badly. Her mother comes into the room and says nothing,
but merely points to the child’s room. The pointing is an example of
________________________.
4. You see a pencil on a desk and say “pencil.” Someone overhears you and hands
you one of their pencils. You say , “No, thank you.” Your saying “pencil” is an example of
________________________.
5. You haven’t eaten in several hours. You say to your host, “I haven’t eaten in
hours.” Your friend says, “Yes, it's been a long time.” You don’t smile but repeat the
statement as you look towards the kitchen. “I haven’t eaten in several hours” is a
________________________.
6. You say “Chan Ho Park” after someone says, “He pitches for the Rangers.”
________________________
7. You round the third base corner without breaking stride as a result of seeing the
third base coach’s hand signal. ________________________
8. You write “Gabe Kapler” after the coach says, “And batting eighth will be Gabe
Kapler.” ________________________
9. At a critical moment in the game you signal the crowd to quiet down so that it will
be possible to hear the coach’s instructions. ________________________
10. A baseball is coming toward you. You say “BALL!!” ________________________
42
A COMPARISON OF THE MAND AND THE TACT
The first basis for comparison concerns who benefits from the response. In the case of the
tact, the listener benefits, because the speaker has provided possibly useful information about
the environment, to which the listener may not have ready access. In the case of the mand,
the speaker is the one who benefits, because the listener provides the reinforcement that
characteristically follows that particular form of verbal response. It could be said that the
speaker also benefits when a tact relation occurs, because the listener provides him with
generalized conditioned reinforcement. But this benefit requires little expenditure of effort on
the part of the listener and is probably not nearly as valuable as the information the listener
receives.
A second basis for comparison concerns the type of controlling variable. The controlling
variable for the tact is a prior nonverbal stimulus. The controlling variable for the mand is a
prior establishing operation. It could be said that the tact tells the listener something about
the environment regardless of the condition of the speaker; whereas, the mand tells
something about the condition of the speaker regardless of the surrounding circumstances.
43
2-8 THE AUDIENCE
Definition: The AUDIENCE is a type of controlling variable with the following features:
DEFINING:
1. The audience is usually a listener in whose presence verbal behavior is typically
reinforced.
2. An audience often controls a large group of response forms.
IRRELEVANT:
1. Whether or not the listener provides reinforcement for the current response
2. The size or specific nature of the group of response forms controlled
In each of the preceding seven elementary relations, the form of the response is specifically
determined by the controlling variable. However, in almost all cases, each of the different
types of controlling variables requires an additional environmental feature to be present
before a verbal response is emitted. The additional feature is a listener. Since the
reinforcement for verbal behavior is mediated by listeners, the presence of a listener is a
discriminative stimulus for verbal behavior in general. In other words, the presence of a
listener sets the occasion for the speaker to obtain reinforcement for speaking. A more
complex analysis arises when the speaker becomes his own listener, and that situation will not
be dealt with in this introduction to the audience variable. Compared to the other types of
controlling variables we have examined, the audience controls a large group of responses
rather than a specific response form. This is because the same audience may be present in a
wide variety of situations in which virtually all of the other types of controlling variables are
also present.
The audience has three different types of control. First, if a stimulus, such as a dog, evokes
only the response “dog” in a given person's repertoire, then the presence or absence of the
audience determines the occurrence of the response in an all or none fashion. If, however, the
same stimulus tends to evoke either the response “dog” or the response “canine,” then the
audience may also determine which of the response forms is emitted. In the presence of a
friend, the less technical form may result: “That animal is a dog.” However, in the presence of
a zoology instructor, the response, “That animal is a canine,” may instead be evoked, even
though the controlling variable is the same dog that evoked the response “dog” in the
presence of the friend.
The final audience effect is to select what is talked about. You may talk extensively about your
most recent love interests in the presence of your peers after class, but when your mother
calls on the phone later that day, you talk almost exclusively about how hard you studied for
your exams and how difficult they were.
The fact that you say “table” in the presence of an English-speaking listener and “la mesa” in
the presence of a Spanish-speaking listener illustrates the audience variable. The fact that you
44
say “table” in the presence of a table and “chair” in the presence of a chair illustrates the tact
relation rather than audience control, although the presence of an audience may determine to
a large extent whether or not the response occurs at this time. Similarly, you may say "table"
when the numerical table your friend has in his backpack would help you complete your
assignment. The presence of your friend (audience) makes it likely that this mand relation will
occur -- that the verbal response "table" will be emitted as a result of the establishing
operation.
Things other than listeners can come to exert audience control. Places such as churches and
libraries may act as “negative” audiences for verbal behavior, because they have been the
occasion upon which many forms of verbal behavior have been followed by
punishment. Listeners can have the same type of effect: we tend not to say certain things in
the presence of listeners who have punished similar responses in the past, or even in the
presence of listeners who are similar to listeners who have punished certain types of verbal
behavior in the past. Places where many verbal relations have often been reinforced may
increase the current likelihood of many types of verbal behavior. The specific things said in
these places will depend on other variables.
The inside of a theater is a negative audience, and the level of verbal behavior, even when
there is nothing on the stage or screen, is usually considerably lower than in the lobby. We see
an increase in the amount of verbal behavior occurring as people enter the lobby after seeing a
movie or play.
2-8 STUDY FRAMES
Answer each of the following by writing either AUDIENCE or the name of one of the other
seven elementary relations. Then record your answers by doing the activity for this Study
Frame Set on the CD.
1. A very handsome man comes into the room, and Margaret says “Quien es el
hombre?” to Sarita. Sarita shrugs her shoulders, and Margaret says “Who is that man?” to
Todd. _________________________
2. Matthew says “tall, dark and gorgeous” when he sees Margaret, and he says
“lithe, blond and charming” when he sees Mary. _________________________
3. You say “hooker” to your friend, in discussing the character in a movie you saw,
and you say ”lady of the night” in describing the same character to your great
aunt. _________________________
4. Calling an acquaintance “Bo” when speaking to him, and calling him “Bozo” when
talking to someone else. _________________________
5. Saying “water” when you have had nothing to drink for a long time, and saying
“bread” when you have had nothing to eat for a long time. _________________________
45
6. Calling someone “hefty” when your 300 pound friend is present, and calling the
same person “huge” when your 85 pound friend is present. _________________________
7. Talking a lot about what a terrible class you are taking, but you suddenly stop
talking about the class when the professor walks into the room. _________________________
8. Saying “Shut off the alarm, please” when an alarm clock is ringing, and saying
“Hand me the umbrella,” when it is raining. _________________________
9. A smoker saying “Can I bum a cigarette?” when he hasn’t had one in six hours, and
saying “Would you mind putting that out?” when someone lights a cigarette in his living room.
_________________________
10. Saying “Those oranges smell delicious” at a roadside fruit stand, and saying “I love
the smell of pine” at a roadside Christmas tree stand. _________________________
11. Saying “Our E.T.A. is eighteen hundred hours” to an armed services co-worker, and
saying “We’ll get there about 6 PM” to a civilian friend. _________________________
12. While playing with a pink pony toy and a green pony toy, saying “I like the pink
pony the best” to your mother, and then saying the same thing to your playmate at school
while playing with the same toys. _________________________
46
VERBAL BEHAVIOR in OTHER RESPONSE MODALITIES
When B. F. Skinner wrote Verbal Behavior, most of his examples were of vocal behavior and
there was considerable discussion of writing. However, Skinner made it clear that the
definition of verbal behavior does not specify the response modality at all. Modalities such as
signing, Morse code, typing, touching Braille, etc. are clearly forms of verbal behavior and can,
therefore, result in verbal stimuli. All of these forms of verbal behavior can easily be classified
as participating in the elementary verbal relations. In fact, in three of the elementary
relations, the form of the response is unspecified and there is no need to discuss analogous
categories. These three are the intraverbal, tact, and mand.
The appearance of a particular response form in American Sign Language (ASL) can be a
function of someone else making the same sign (ASL echoic). There is a written form of ASL
(developed by Dr. William Stokoe) that is somewhat analogous to a phonemic form of written
English. Rather than the letter corresponding to speech sounds, the “letter” corresponds to
significant features of the signs and, therefore, has point-to-point correspondence that allows
for relations analogous to textural behavior, copying a text, and taking dictation. Although
these relations are possible, they have not become commonplace in the deaf community.
A sign in ASL as the result of word written in English is not analogous to textural behavior,
because there is no point-to-point correspondence between the sign and the controlling
stimulus. This would have to be classified as a type of ASL intraverbal. Another form of ASL
intraverbal would involve making the sign “cat” as a result of seeing the sign “dog.” The ASL
intraverbal that has the English word as the controlling variable is the same type of intraverbal
discussed in the acquisition of a foreign language. Indeed, ASL and English are two different
languages. The analysis for the other forms is similar. The key point to remember is that
Skinner’s analysis of verbal behavior is not based upon the type of behavior involved, but
rather upon the way reinforcement is generated for operant behavior designated as verbal.
The reinforcement is mediated by a listener.
ASL has often been regarded by individuals with a traditional approach to language as less than
a complete system of verbal behavior. To a great extent this is because language has been
defined by analyzing the characteristics of vocal language systems such as English. Skinner’s
analysis does not depend upon the type of response; rather, it is an analysis of relations
between certain types of controlling variables and responses involving any muscles, not just
those involved in vocalization. When analyzed in this way, both English and ASL are complete
systems of verbal behavior.
47
CHAPTER 3: EXTENSIONS OF VERBAL BEHAVIOR The verbal relationships we have studied thus far represent well-established responses
controlled by specific stimuli. Verbal behavior, however, is not so simple. A major complexity
involves verbal behavior that occurs in the presence of novel stimuli. B. F. Skinner, in Verbal
Behavior, opens the discussion of the “extended tact” by writing:
But a verbal repertoire is not like a passenger list on a ship or plane, in which one name
corresponds to one person with no one omitted or named twice. Stimulus control is by no
means so precise. If a response is reinforced upon a given occasion or class of occasions, any
feature of that occasion or common to that class appears to gain some measure of control. A
novel stimulus possessing one such feature may evoke a response. There are several ways in
which a novel stimulus may resemble a stimulus previously present when a response was
reinforced, and hence there are several types of what we may call “extended tacts.”
Extended tacts are the extension of stimulus control over verbal responses to novel nonverbal
stimuli. We will discuss three ways in which control extends to novel nonverbal stimuli, calling
them generic extension, metaphorical extension, and metonymical extension. There are also
several kinds of extension of control over verbal responses acquired in mand relations and we
will briefly describe these. In this chapter, we will also introduce four methods used by the
verbal community to teach speakers tact relations in which the nonverbal stimulus is private,
or covert.
48
STIMULUS FEATURES
We talk about behavior being under the discriminative control of a stimulus. Most often we
talk as if a stimulus is some object or event, such as a dog or an explosion. But the stimulus
that has discriminative control over any particular verbal response may include more of the
environment, or less, than that part of the world we view as a single object or event. Any
recognizable object or event is part of a larger environment, and the object together with the
larger environment may be what controls a particular response. For example, a dog behind a
fence may be responded to differently (verbally or nonverbally) than the same dog outside the
fence. The tact relation of saying “safe” in the presence of the dog behind the fence is a
different tact from that of saying “dangerous” in the presence of the same dog loose. Both
verbal responses occur in the presence of the same dog, but other features of the situation,
together with the dog, determine which response occurs. Parts of objects or events also can
acquire discriminative control over verbal behavior. We learn the tacts “tail,” “ears,” and
“teeth” as verbal responses under discriminative control of particular parts of dogs (and other
creatures).
All objects or events, and their components or parts, have in common certain physical
properties. Some properties common to all objects are length, width, height, hue, location,
mass. Some properties common to all events are duration, latency, and intensity. These
physical properties can enter into tact relations, too. We respond to various values of the
property of hue by calling some dogs “black” and some “brown.” These kinds of tacts are
called “abstract” because, in the everyday world, a single stimulus property cannot be present
without other properties being present as well.
The multiple elements of a complex situation, the parts of complex stimulus objects and
events, and their stimulus properties, are called stimulus features. As Skinner pointed out in
his remarks about tact extension, any feature of an occasion can gain some control over a
response that is reinforced on that occasion. However, it is generally the case that effective
responding is controlled by only one or a few of the features present at any given time. If a
pigeon gets food only if it pecks a key when a keylight is red (as opposed to other colors), then
the pigeon behaves most effectively when only the property of hue differentially controls
pecking – whatever the values of other properties of the keylight. If the pigeon pecks at bright
red keylights but not at dim red lights, then the property of intensity has gained unwanted
control over the response. Likewise, if the pigeon pecks a red keylight only when it is in the
center of a wall, but not when it is located to the left or right of center, then control by
location is interfering with maximally effective responding.
In teaching humans the verbal relations needed to get along in their verbal community, we
reinforce saying “dog” (but not “cat” or “parrot”) on various occasions when a dog is present,
but not when only a cat or other animal is present. After we teach tact relations in which
common objects and events function as discriminative stimuli, we begin to teach the more
subtle tacts, in which parts of those objects come to function as the discriminative stimuli
(“tail,” “head,” etc). We also teach verbal responses under discriminative control of the
stimulus properties of objects or events – responses such as “red,” “loud,” and “tall.” Finally,
49
relations between the objects and other objects also become discriminative stimuli in yet more
complex tact relations (e.g., “behind,” “under” “to the left”).
3-1 EXTENDED TACTS
As you have learned, the discriminative stimulus in verbal relations called “tacts”
is nonverbal. The stimulation that controls the verbal response in a tact may be objects,
events, their parts, relations among them, or an abstract property of an object or an
event. Any occasion on which a verbal response is reinforced has many features, and any one
or combination of those features may be the variable that functions as the controlling stimulus
for the verbal response in the tact. When the reinforcing practices of the verbal community
are considered, one can classify the stimulus features of any object or event as
either relevant orirrelevant.
A RELEVANT stimulus feature may be defined as:
Any feature of an occasion that is required to be present when a response form is typically
reinforced by listeners of a verbal community.
An IRRELEVANT stimulus feature may be defined as:
Any feature of an occasion that is (or may be) present when a response is typically reinforced
by listeners of a verbal community, but upon which reinforcement is not contingent.
Let’s consider the tact “dog” – that is, emission of the verbal response “dog” under stimulus
control of a dog. Dogs have several characteristic stimulus features, including canine teeth,
four legs, claws on feet, tails, fur, barking, color, size, shape, and so on. Which of these
features must be present before our verbal community typically reinforces saying “dog”? (For
introductory purposes, it will be helpful to ignore strange examples or uncommon exceptions:
e.g., a Basenji doesn’t bark, but is still called a dog.) Fur would typically be required (ignoring
the Mexican hairless Chihuahua); also barking, four legs, canine teeth, claws, tails (cropped or
not), and perhaps others.
Would we require a specific color? A specific size? Since neither the size nor the color would
make a difference to whether the verbal community reinforces the tact “dog,” the size and
color of the stimulus object is irrelevant. The features that we would require to be present are
relevant. A horse is an improper occasion for saying “dog,” partially because it has hooves
rather than claws. It also has the wrong kind of teeth. Note that although the size of a horse is
different from most dogs, it is still irrelevant. In the presence of a Saint Bernard as big as a
horse, the verbal community still reinforces the verbal response “dog.” A chicken is even less
likely to be the occasion for saying “dog,” because it shares very few features in common with
a dog. On the other hand, a wolf has almost all of the relevant features of a dog, and we
would not be at all surprised to find someone calling a wolf a dog. We would not be quick to
punish this response, whereas we might be in the case of the chicken.
Important to note is that the verbal community is not perfect in its reinforcing practices and
there is no way for perfection to be achieved. That is because the world is a continuous flow
of activity – of changing objects and events – and verbal responses under control of those
50
activities are, to be useful, either/or kinds of things. Either the verbal response is “dog” or it is
not (with minor variations allowed), whereas the nonverbal stimuli that are present when that
response is reinforced vary tremendously from one occasion to the next. This means that the
rather precise verbal response “dog” may be reinforced in the presence of a huge variety of
objects, with no end to the future possibilities. So on the response side of any tact relation,
great precision is required; but on the stimulus side, great variation is allowed (and even
encouraged). Nevertheless, there are limits to the stimulus variation allowed, or speakers and
listeners would be unable to affect one another’s behavior.
When the stimulus event in a tact relation is a behavioral event, the practices of the verbal
community are often much less precise than when objects and their parts are the stimuli in
tact relations. For example, if we observe a child and say he is “doing math homework,” other
members of the verbal community may disagree as to whether that is what the child is
doing. This is the reason behavior analysts develop behavioral definitions, which are
specifications of the relevant and irrelevant features of the tact “doing homework.” Behavior
analysts want to insure that all the observers collecting the data are engaging in verbal
responses under control of the same features of the complex situation they are observing. The
behavioral definition specifies the relevant and irrelevant stimulus features of the behavior the
observer is counting. It is probably beneficial to the learner if the definition of the behavior to
be learned reflects the practices of the larger verbal community as closely as possible.
51
3-1 STUDY FRAMES
Write RELEVANT, IRRELEVANT, or NEITHER in the blanks. Then record your answers by doing
the activity for this Study Frame Set on the CD.
Specify the following features as relevant or irrelevant for the tact “triangle.”
1. number of sides _________________________________________________
2. size of the figure _________________________________________________
3. color of the object presented _______________________________________
4. closed or open figure _____________________________________________
Specify the following features as relevant or irrelevant for the tact “chair.”
5. number of seats _________________________________________________
6. presence or absence of a back ______________________________________
7. color __________________________________________________________
8. material ________________________________________________________
9. size ___________________________________________________________
Specify the following features as relevant or irrelevant for the tact “watching TV”
10. volume is turned up on TV___________________________
11. TV program is in a language the watcher speaks _____________________
12. TV is turned on ______________
13. channels are being changed ___________________
14. watcher is looking at the TV screen ________________
Specify the following features as relevant or irrelevant for the tact “brushing teeth.
15. holding toothbrush _______________
16. time of day ___________________
52
17. whether water is running or not _________________
18. moving brush against teeth __________________
19. rinsing mouth _______________
53
TYPES OF EXTENDED TACTS
Skinner restricted his treatment of extension almost entirely to the tact relationship, and that
is perhaps the relationship whose extensions are most interesting. Skinner also discussed
mand extension, a topic that is treated separately here because the mand can be somewhat
free from stimulus control.
An extension of a tact relation occurs when a novel nonverbal stimulus controls a response
that has been reinforced in the presence of other stimuli that share some features with the
novel stimulus. A new situation (novel stimulus) may have varying degrees of resemblance to
an old situation. The degree of resemblance depends on the number of relevant and the
number of irrelevant features that characterize the novel
stimulus. (Remember, relevant features are stimulus features that are the basis for the verbal
community’s reinforcement of the response form.) This fact allows for three general
categories of tact extension based upon the degree to which a novel stimulus shares relevant
or irrelevant features with a stimulus that has already gained some measure of control over a
particular verbal response as a result of a reinforcement history.
54
3-2 GENERIC EXTENSION
Definition: GENERIC EXTENSION is an instance of verbal behavior that has the following
features:
DEFINING
1. The response form must be a previously learned one.
2. The stimulus situation must be novel.
3. The novel stimulus must have all of the features that are relevant to the verbal
community’s reinforcement practices.
IRRELEVANT
1. Formal or dynamic characteristics of the response
2. Formal or dynamic characteristics of the stimulus
3. Whether or not the current response is followed by reinforcement
Generic extension results in classifying things into groups or categories in accordance with the
reinforcing practices of the speaker’s verbal community. For example, if a child has been
taught to say “dog” in the presence of (under stimulus control of) the family Cocker Spaniel
and also in the presence of her uncle’s Beagle, any tendency to say “dog” in the presence of a
Labrador or a Sheltie would be a generic extension of the previously acquired tact relationship
(and the reinforcing community would reinforce by agreement or approval: “Yes, that’s a dog –
like Pooh”).
Once a response has been reinforced in the presence of a novel stimulus, the process of
extension is not required to explain the occurrence of the same response in the presence of
that stimulus at some time in the future. The reinforcement has resulted in broadening the
class of objects that evoke the verbal response “dog” in this child’s repertoire. Extension
requires a novel stimulus; therefore, if the child sees the same German Shepard the next day
and again says “dog,” that is not generic extension, because the stimulus is not novel on this
occasion. If a child says “dog” in the presence of a badger, this is not generic extension,
because the novel stimulus (badger) does not have all of the relevant features necessary for
the verbal community to provide reinforcement for the response “dog.” This type of extension
is presented next.
3-2 STUDY FRAMES
Fill in the blank by writing either GENERIC or NONGENERIC, or NOT AN EXTENSION. Then
record your answers by doing the activity for this Study Frame Set on the CD.
55
1. A child learns to say “chair” in the presence of a straight-backed, wooden chair
and then later says “chair” in the presence of a new large, rounded, overstuffed chair.
_____________________________
2. When someone learns to say “guitar” in the presence of a guitar and then later
sees a banjo for the first time and says “guitar,” that response is an example of
___________________
3. The first time a child goes to the zoo, he learns to make the response “zebra” in
the presence of a zebra. ________________________________
4. The same child sees an animal (an aardvark) and asks his father what the name of
that animal is. His father says “aardvark” and then as a result, the child says “aardvark.” The
child’s response is ________________________________.
5. You have learned to say “12-point” when you see the words you are now reading
in (12 pt) Tahoma typeface. You later say “12-point” when you see the words you are now
reading in (12 pt) Arial typeface. Your response is an example of ________________.
6. A therapist learns to say “consequence” under control of the delivery of
candy following a child’s correct tact. Then she says “consequence” under control of the
delivery of a teacher’s “try again” that follows a child’s incorrect tact. The therapist response
“consequence” in the new situation is _________________________.
7. A graduate student learns to write “+” on a “tact data sheet” if a child’s response
is a correct tact. The student then writes “+” on a “mand data sheet” if a child’s response is a
correct mand. Writing “+” on the mand data sheet is ___________________________.
8. A child sees a billboard that reads, “Pokemon: The Movie.” At the dinner table
that night the child says, “Can we go to see Pokemon?” _______________________
9. A child is given a can of cola at home and told that it is called “pop.” Later, when
he drinks a fountain drink at a fast food store, the child says “pop.”
________________________
10. While in class, a computer programmer learns to call text that follows the symbol ‘
a “comment.” Later, while investigating another computer program, she sees the symbol ‘
followed by some text and says, “That is a comment.” ________________________
11. A child is taught to say “ball” in the presence of a baseball. Later, when the child
sees a hockey puck, he says “ball.” __________________________
12. A child learns to say “danger” when he sees a small picture of a skull and
crossbones on a bottle of bleach. Later, when he sees a large sign with skull and crossbones on
the door of a building, he says to his sister “danger.” _______________________________.
56
3-3 METAPHORICAL EXTENSION
Definition: METAPHORICAL EXTENSION is a form of verbal behavior with the following
features:
DEFINING
1. The response form has previously been acquired.
2. The stimulus situation is novel.
3. The novel stimulus has some, but not all, of the relevant features of the stimulus
that previously controlled the response.
IRRELEVANT
1. Formal or dynamic characteristics of the response
2. Formal or dynamic characteristics of the stimulus
3. Whether or not the current response is followed by reinforcement
If someone had acquired the response “dog” in the presence of several different types of dogs
and then saw a wolf for the first time and said “dog,” that would be an example of
metaphorical extension. The response form is controlled by the many features a wolf has in
common with dogs. To say “dog” when seeing a wolf is quite likely; to say “dog” when seeing
a cat for the first time is not so likely, but it would still be a metaphorical extension. This is
because the wolf has many of the relevant features required by the verbal community, but a
cat has only a few of the relevant features required for the verbal community to reinforce the
response “dog.” The fewer the relevant features of a novel stimulus, the less likely the
response is to be emitted in the presence of that novel stimulus. Also, striking differences in
irrelevant stimulus features, even if several relevant features are present, will result in the
probability of the response being quite low.
Metaphorical extension is often considered to be “creative.” It seems likely that the reason for
calling a metaphorical extension “creative” is that the probability of the response under those
conditions is low, therefore the response under those conditions is “original” (i.e.,
“creative”). The fewer the relevant features present, or the more striking the irrelevant
features are, the more “creative” the extension is likely to be viewed. Even so, the verbal
community must be able to ascertain that the stimulus features of the novel stimulus have
something in common with the stimulus features of the conditions that are ordinarily present
when the verbal response is reinforced.
We have already seen that if all relevant features are present in a novel stimulus, then the type
of extension is generic. Because any feature of an occasion can gain some control over a
response made on that occasion, it is even possible for a response to be evoked
when only irrelevant features that accompanied relevant features are present in a novel
situation. This latter type of extension is presented in the next section.
57
3-3 STUDY FRAMES
Write GENERIC, METAPHORICAL, or NEITHER in the blank after each example. Then record
your answers by doing the activity for this Study Frame Set on the CD.
1. You have learned to say “espadrille” under control of a sandal with a canvas upper
and roped sole. Now someone shows you footwear that is a sandal with a canvas upper and a
leather sole. You say, “That’s an espadrille.” _______________________________
2. You have learned to say “Mozart” as a result of having heard several of his piano
concertos. Now, while listening to a piano concerto of a totally different style by Wagner, you
say “Mozart.” ___________________________________
3. Having learned to say “tree” in the presence of oaks and maples, you now see a
white pine for the first time and say “tree.” ______________________________
4. Having learned to say “tree” in the presence of oaks and maples, you see a lilac
bush for the first time and say “tree.” _________________________________
5. The verbal community has reinforced a child’s saying “elephant” in the presence
of small stuffed toys with trunks, small tails, etc. The child goes to the zoo and, seeing a living
elephant for the first time, says “Elephant!” __________________________.
6. The verbal community has reinforced a child’s saying “elephant” in the presence
of small stuffed toys with trunks, etc. The child’s dad bends down and swings his arm in front
of his face, with his index finger crooked. The child says “Elephant!!”
___________________________.
7. You look out of the plane on a very dark night and see the colored lights of a city
scattered against the blackness of the earth. You say “jewels on
velvet.” ______________________________
8. A child sees a purple dragon on TV and asks, “What’s that?” His mother says, “It’s
a dragon,” to which he replies, “ Dragon.” Later in his brother’s room he sees a green dragon
on a wall calendar and asks, “Is that a dragon?” to which his brother replies “yes.” Days later
at a museum showcasing Asian art, he sees a statue of a black dragon and says “A dragon!”
____________
9. You acquire the response, “butterfly” in the presence of a range of winged
insects. Then one day you see a cloud in the shape of a butterfly and say, “It’s a butterfly.”
_________
10. A child is taught to say “towel” in the presence of a white towel. Later, in the
presence of a white car, the child says “towel.” __________________________
58
11. While on summer vacation at an aunt’s farm, a three-year-old girl learns to call a
black and white yearling a “calf.” A few days later, the little girl sees the black and white
yearling and says “calf.” ______________________
12. A little boy sees Army personnel in uniform on TV and his fatehr tells him they are
"soilders". Later he sees some Boy Scouts in uniform and calls them
"soldiers". ______________________
59
3-4 METONYMICAL EXTENSION
Definition: METONYMICAL EXTENSION is a form of verbal behavior with the following
features:
DEFINING
1. The response has already been acquired in one or more of the elementary verbal
relationships.
2. The stimulus situation is novel.
3. The novel stimulus must have none of the relevant stimulus features of the class
of stimuli that previously controlled the response.
IRRELEVANT
1. Formal or dynamic characteristics of the response
2. Formal or dynamic characteristics of the stimulus
3. Whether or not the current response is followed by reinforcement
Because any feature of an occasion may gain some control over a response reinforced on that
occasion, it is possible for irrelevant stimulus features to gain control of a response. For
example, it is often reported by the press that “a statement was issued by the White
House.” Because this usage is reinforced now by the verbal community, this is not an example
of a metonymical extension. However, the first time such a statement was made was an
example of a metonymical extension. Skinner says this type of extension occurs because the
controlling stimulus for the novel response frequently accompanies the relevant features.
Sometimes a response occurs in a situation where there appears to be no appropriate
stimulus present. For example, a child who always saw flowers on the breakfast table may sit
down one morning when there are no flowers on the table and say “flowers” (probably with a
“no” in front of the “flowers”). Responses are not controlled by the absence of something. We
must, therefore, find some aspect of the current situation that is controlling the
response. Skinner accounts for the tendency to say “flowers” by suggesting that all of the
other features of the breakfast table that have been present when the response “flowers” was
reinforced before (when there were flowers on the table) can also gain some control over the
response “flowers.”
Saying “stop” as the result of seeing a red hexagonal sign with no letters on it is another
example of metonymical extension. The shape and color are irrelevant stimulus features that
frequently accompany the written word “stop.” If you had learned to say “espadrille” as a
result of seeing a sandal with a canvas upper and a roped sole with a specific trademark inside,
you might have some tendency to say “espadrille” as a result of seeing a shoe with a leather
upper and leather sole with that same trademark. In this case, the irrelevant feature of the
trademark exerts some control over the response.
60
61
3-4 STUDY FRAMES
Write GENERIC, METAPHORICAL, METONYMICAL, or NONE in the blank following each
example below. Then record your answers by doing the activity for this Study Frame Set on the
CD.
1. A child learns to say “CD” when a CD is inserted in a CD player. He later sees the
CD player alone and says “CD.” _________________________________
2. When we teach a child to say “red” by showing him several red circles, he may
later say “red” when he sees a blue circle. _______________________________
3. A child learns that the object on which his food is served in the dining room is a
table. One morning at breakfast, he sits down at the same place and says that he is going to
eat at the “table.” _____________________________________
4. A chef learns to say he is “making a cake” when he places butter, then eggs, then
sugar, then flour into a pan and stirs. When later he places eggs, then sugar, then butter, then
flour into a pan and stirs, he says he is “making a cake.”_______________________
5. Skinner learned to say “reflex” in the presence of an observed relation between
an antecedent stimulus and a subsequent response. He then said “reflex” in the presence of
an observed relation between a response and a subsequent stimulus. (Later, he suggested
calling the second kind of relation a “reflex” was a mistake.)
6. A student learned to call boxes, cartons, jars, and cups with items inside them
“containers.” He then said “container” when he opened a book and saw words inside
it.___________________________
7. A child is told “that’s the phone” when the phone rings. Later when the phone
rings the child says “phone.” _______________________
8. You learn to call the taste of a hot, dark brown, unsweetened beverage
“coffee.” Later you take a bite of light brown ice cream and say
“coffee.” ________________________
9. An adult points to a picture of a skull and crossbones on a bottle of bleach and
tells a child that this means “danger.” The child says “danger” when the adult then shows him
the bottle with the skull and crossbones on it. He later sees a similarly shaped bottle of water
and says “danger.” ____________
10. A person on vacation in Hawaii sees a Hawaiian point to a necklace composed of
fresh orchids and say “lei.” After returning home, the vacationer makes a necklace of plastic
flowers and tells her daughter it is a “lei.” ______________________
11. A young girl is taught to say “comics” in the presence of her big sister’s collection of
Archie comics. When visiting a neighbor, the little girl sees a book of oil paintings from the
Metropolitan Museum of Art and says “comics.” ______________________________
62
12. Any time a postal worker comes to the door, Hal’s behavior of saying “mailman”
has been reinforced. Then one day a missionary comes to the door and Hal says “mailman.”
______________________
63
3-5 WAYS OF TEACHING RESPONSES TO PRIVATE STIMULI
There are four major ways in which the verbal community, in trying to teach verbal responses
to private stimuli, attempts to overcome the difficulty of not having direct access to the
stimulation that must acquire control over particular response forms. Following Skinner, we
will call them the methods of 1) public accompaniment; 2) collateral responses; 3) common
properties; and 4) response reduction.
(1) PUBLIC ACCOMPANIMENT
Definition: PUBLIC ACCOMPANIMENT is a method by which the verbal community brings
verbal behavior under the control of private stimuli, and it has the following features:
DEFINING
1. A private stimulus, which is often stimulation of free nerve cells
2. Is reliably accompanied by a collateral public stimulus
3. Reinforcement for a given response is contingent upon the presence of the public
stimulus
IRRELEVANT
1. The type of private stimulus
2. Whether or not there is any causal relationship between the public and private
stimulus (they need only be correlated)
3. Whether or not the current response is reinforced
If we see an object strike someone, or see red or swollen tissue, we assume that these visual
stimuli are correlated with private stimulation of free nerve endings (pain receptors). If we see
a rash on someone’s arm, we assume that the rash is correlated with the private stimulation
we call itching. We can teach tact relations in which private stimulation is discriminative when
the verbal community has access to a public accompaniment of the private stimulation. An
example of this is a parent who sees a child hit his head against the table. The parent says,
“That hurts,” while the child is feeling the painful stimulation. In this way, the child learns to
say “that hurts” when painful stimulation is present. Note that the parent’s pain receptors
were not stimulated, but his visual receptors were stimulated by the public stimulus (the
child’s head hitting the table) that accompanied the pain felt by the child.
When members of the verbal community are being stimulated by public events while a learner
is concurrently being stimulated by private stimulating events, the verbal community can teach
the learner to emit a particular form of verbal response under control of the private
stimulation that is typically present on such occasions. The learner’s verbal response “hurt”
(or “itch”) can be brought under discriminative control of the private stimulation if that
stimulation reliably has a public accompaniment that can function as a discriminative stimulus
for the teaching of the appropriate verbal response (e.g., “that hurts”).
64
Of course, not all public stimuli have collateral private stimuli. A rash may not, in fact, itch. An
area of swollen or bruised skin may not hurt. On the other hand, a person may feel an itch
although there is no public collateral stimulus, or a person may have pain without any public
accompaniment. It is also possible that a person may sometimes say “that hurts” or “that
itches” when there is no painful stimulation or no itch. This is one reason that people tend to
mistrust reports of private stimulation. There is no way for the listener to verify the presence
of the stimulation that the speaker claims is present.
65
COLLATERAL RESPONSE
Sometimes the verbal community takes its cue from a collateral response of the learner. The
learner’s private stimulation (for example, of free nerve endings) may be accompanied by
crying or moaning. A toothache or stomach ache may have the collateral response of holding
one’s jaw or stomach. A full bladder may be accompanied by wriggling, etc.
Definition: The method of teaching responses to private stimuli by COLLATERAL
RESPONSES has the following features:
DEFINING
1. A private stimulus
2. Some collateral behavior of the person who is privately stimulated
3. The community’s reinforcement is contingent upon the presence of the collateral
responses.
IRRELEVANT
1. The type of stimuli
2. The form of behavior
3. Whether or not the current response is reinforced
By response, we mean some form of behavior involving muscle or gland activity. Bleeding
cannot be a collateral response, but it can be a public accompaniment. Looking pale is also not
a collateral response, although the constriction of blood vessels could be considered a
collateral response that has looking pale as a response-product. The occurrence of the
collateral response is the occasion on which the verbal community teaches the learner to say
“hurt” or “pain.” Because stimulation of pain receptors (free nerve endings) is present on
most occasions when collateral responses cue the verbal community to teach the learner
responses like “that hurts,” the pain can become the stimulus that controls the response
“hurts.”
66
3-5 STUDY FRAMES
Write PUBLIC ACCOMPANIMENT, COLLATERAL RESPONSE, or NEITHER in the blank. Then
record your answers by doing the activity for this Study Frame Set on the CD.
1. You tell a child “that hurts” when you see bruised skin on his arm.
___________________________
2. You reinforce a child’s saying “it hurts” when you observe her grimacing.
___________________________
3. You reinforce a child’s saying “it hurts” when you observe the child screaming.
________________________
4. An empty cookie jar (that was full when you left the room a short while ago) is a
discriminative stimulus for you to say “You must have a full stomach.”
________________________________________
5. A child, having learned to say “pounding” under a variety of circumstances, now
says, “I have a pounding headache.” __________________________________
6. You say to your friend, “You must be bored,” after seeing him twiddling his
thumbs for twenty minutes. ___________________________
7. You see a large stack of papers on your roommate’s desk and say, “Feeling
stressed?” _________________________________
8. On Halloween your child sees someone dressed up like a clown and giggles. You
say “You think he’s funny.” __________________________
9. Marnie is watching a movie and screams when the ghoul appears at the heroine’s
door. Her mom says, “Are you scared, sweetie?”_____________________
10. You see a group of children, all but one of whom are eating ice cream cones. You
say to the child without ice cream, “You’re feeling jealous, aren’t
you?”____________________
67
68
3-6 MORE WAYS OF TEACHING RESPONSES TO PRIVATE
STIMULI
(3) COMMON PROPERTIES and (4) RESPONSE REDUCTION
There are two more ways that the verbal community can teach speakers tacts in which
the discriminative stimuli are physical but private, or inaccessible to the members of the verbal
community. The first method is called common properties.
Some of the properties of a public stimulus may also be present in a private stimulus. The
types of properties that seem frequently to be shared by private and public stimuli are those
that are geometrical, temporal, orintensive. When private stimuli have such properties and
the individual has already learned to respond to these properties when they are features of a
public stimulus, the response is likely to be extended to the private stimulus. This type of
extension can be any of the three categories that have already been discussed: generic,
metaphorical, or metonymical.
Definition: The method of learning to respond to private stimuli known as COMMON
PROPERTIES has the following features:
DEFINING
1. A learned response topography controlled by a feature or features of a public
stimulus
2. A novel private stimulus that has properties in common with the public stimulus
IRRELEVANT
1. Whether the properties shared by the private stimulus and that public stimulus
are relevant or irrelevant to the reinforcing practices of the verbal community
2. Type of extension
3. Type of properties
4. Whether or not the current response is reinforced
If you have learned to call a public stimulus that oscillates in intensity “throbbing,” then there
is a good likelihood that a headache that oscillates in intensity may also evoke the response
“throbbing.” This is an example of generic extension, since the pain has all of the relevant
features of the public stimulus present when the verbal community reinforces the response
“throbbing.”
The last way a speaker can acquire a tact in which a verbal response is under stimulus control
of private stimulation is called response reduction. By this method, we learn to respond to
69
private stimulation generated by our own behavior. The verbal community can teach us to
describe what we are doing when members of the community observe us engaging in various
kinds of action. The verbal community teaches us to describe our acts as “speaking,”
“walking,” “swearing,” “kissing,” etc. We learn to describe our own behavior because the
verbal community can observe our public actions and differentially reinforce verbal response
forms that are appropriate to our various acts. At the same time we learn to observe those
public events, private kinesthetic stimulation is also occurring. So the individual who is
learning to describe his own actions in terms of how they look or sound (both to that individual
and to the verbal community) is also being stimulated by kinesthetic stimuli that are produced
by the body’s movement. If the behavior is reduced in magnitude to the point that there is no
externally observable response-product, the private stimulation can still affect kinesthetic
receptors. Therefore, the speaker could tact the private kinesthetic stimulation as his own
behavior. We may talk more and more quietly until we are producing no easily observable
movement or sound, yet we can report that we are engaging in verbal behavior, because the
kinesthetic stimuli that accompanied the overt behavior gained some control over the verbal
response “I’m talking.”
Definition: RESPONSE REDUCTION has the following features:
DEFINING
1. A verbal response controlled by the speaker’s own overt behavior
2. Private kinesthetic stimuli that are collateral with observable public stimuli
3. The response is reduced in magnitude, producing no public stimuli
IRRELEVANT
1. The topography of the response
2. The particular form of the speaker’s overt behavior
3. Whether or not the current response is reinforced
One general difference in the methods by which speakers learn to respond verbally
under stimulus control of internal stimulation is whether the listeners directly teach the
learners to respond that way or whether the learner’s verbal behavior extends to internal
stimulation as an indirect result of the verbal community’s reinforcement practices. In the
methods of public accompaniment, collateral responses, and response reduction, the private
stimulation and the public stimulation co-occur. Because the public stimulation is present, the
verbal community is directly responsible for the reinforced responding that occurs in the
presence of both the private and public stimulation. On the contrary, the method of common
properties involves tact extensions that occur when the public stimulus is not present. These
verbal responses extend to features of the private stimulation that are similar to features of
public stimulation that have served in the past as discriminative stimuli for verbal responses.
This difference will help you identify examples of the method of common properties.
70
Here is a way to help distinguish among the three methods that involve co-occurrence
of private and public stimulation in the presence of the verbal community teaching the
discrimination. In the method of public accompaniment, the public stimulus is not the actions
of the learner but other publicly available events. In the methods of collateral responses and
response reduction, the learner’s actions (behaviors) are the public stimuli that control the
teacher’s verbal behavior. Notice that the word “responses” is part of the name of both these
methods. The names refer to the learner’s responses that occur at the same time as the
private stimulation. The difference in these last two methods has to do with whether the
private stimulation merely occurs at the same time as the public response of the learner
(collateral responses); or whether the private stimulation actually results from learner’s action
(reduced forms of the muscular and glandular activity that originally controlled the verbal
response).
3-6 STUDY FRAMES
Write PUBLIC ACCOMPANIMENT, COLLATERAL RESPONSE, COMMON PROPERTIES or
RESPONSE REDUCTION in the blank. Then record your answers by doing the activity for this
Study Frame Set on the CD.
1. A person has learned to say “ringing” when the school bell rings and when the
telephone rings. Now that person hears similar sounds in his ear and says his ears are
“ringing.” ____________________
2. When a child sways and stumbles after getting off a roller coaster, her mother
teaches her to say she is dizzy. Later, when a fever causes disturbance of the vestibular system
and the world seems to spin for the child, the child says “I’m
dizzy.” _____________________________
3. A father sees a mosquito bite with scratch marks around it, and he teaches his
child to say “that itches.” The child later gets a fever blister and says “it itches.”
__________________
4. The verbal community teaches a person to say “heavy” when he moves with
ankle weights on. Later, when that person is suffering from arteriosclerosis, he tells his
physician that his legs are “heavy.” _________________________________
5. A boy’s face contorts in a series of short jerking motions. His father says, “You’re
twitching.” Later the muscles of his face twitch ever so slightly and the boy tells his father that
his face is twitching again.______________________
6. A little girl smiles and her mother says, “You seem happy today.” Later, without
smiling, she tells her mother, “I’m happy.”_______________________
71
7. A child looking at an unfocused image from a projector is told that the image is
“blurry.” Later, in describing his own vision after sustaining a concussion, he uses the word
“blurry.”_____________
8. A girl and her grandmother watch a slug move along in their backyard garden. The
grandmother says, “Look at how slowly the slug moves, it can’t move any faster.” Later, when
the girl feels unable to move normally due to illness, she says, “I feel like a
slug.”_____________________
9. After a young boy work on a house of cards for twenty minutes, his baby sister
crawls by and knocks it over. His mother says, “That’s frustrating.” Later, the wind blows his
hat away and he says to his friend, “That’s frustrating.” ________________________
10. A girl is learning to play pool. After missing several shots in a row, she throws
down her stick. Her father says, “You must feel frustrated.” Later, when having difficulty
solving math problems in her classroom, she tells her teacher, “I’m
frustrated.”_________________________
72
CHAPTER 4: MULTIPLE CAUSATION
In Verbal Behavior, Skinner notes: “Two facts emerge from our survey of the basic functional
relations in verbal behavior: (1) the strength of a single response may be, and usually is, a
function of more than one antecedent variable, and (2) a single antecedent variable usually
affects more than one response” (p. 227). These two facts help account for the great variety
of novel verbal relations that constantly appear in the behavior streams of individual speakers
who have acquired a repertoire of tacts, mands, and intraverbals, as well as the point-to-point
correspondence seen in textual, echoic, and transcriptive (taking dictation and copying text)
relations. Once this repertoire of elementary verbal relations has been established through
reinforcement mediated by listeners in the speaker’s verbal community, control by novel
combinations of antecedents over instances of verbal behavior often occurs. Novel response
forms may also emerge when two or more verbal responses are strengthened by a current
state of affairs.
We have evidence of multiple causation of verbal responses when we can point to multiple
sources of control over a speaker’s verbal responses. Literature and humor are rich sources of
evidence for multiple control in verbal behavior. In the following sections, we will examine
both kinds of multiple control. First, we consider ways in which antecedent variables can
combine to produce new control over previously learned response forms. Second, we will
examine ways in which new response forms can result from concurrent control by novel
combinations of antecedent events that individually already control some verbal
responses. So this chapter is about some of the ways that multiple causation works.
4-1 MULTIPLE CONTROLLING VARIABLES
Definition: MULTIPLE CONTROLLING VARIABLES is characterized by a controlling relation
having the following features:
DEFINING
1. There is a single response form.
2. That response is simultaneously strengthened by two or more controlling
variables.
IRRELEVANT
1. Response topography (the particular form of the response)
2. The type of controlling variable (discriminative stimulus or establishing operation)
3. The specific number of multiple variables
4. Whether or not the current response is followed by reinforcement
73
If someone asks you what time it is and it is five o’clock, your verbal response “five o’clock” is
likely controlled by both the visual stimulus of the position of the hands on a clock and by the
prior mand for the correct time. The last word in a line of poetry is often the function of more
than one variable. In the lines from Shakespeare--
“Golden lads and girls all must,
as chimney sweepers, come to dust.”
the phrase “come to dust” is multiply controlled. The poem is about death, and therefore, this
thematic source of control strengthens any response form about death. A second source of
control is the intraverbal relationship between “chimney sweepers” and “dust.” A final
controlling variable is the formal control between the responses “must” and “dust.” It is these
many kinds of multiple control, each of which is “just right,” that results in our calling great
poetry great.
Another significant example of this type of multiple stimulus control involves the effect of
particular audiences on what speakers say. Saying “canine” in the presence of a particular
person may be controlled by the presence of an appropriate animal as well as by the fact that
this audience is a zoologist. A particular audience usually strengthens a particular group of
responses, and a nonverbal or verbal stimulus determines the specific response form. We are
much more likely to say “cardinal” in the presence of both the bird and a friend who is a bird
watcher than we are in the presence of the bird alone. Saying “cardinal” in the presence of the
bird alone is not an example of multiple control.
74
MULTIPLE RESPONSES
Definition: MULTIPLE RESPONSES is a situation that has the following features:
DEFINING
1. There is a single controlling variable.
2. That controlling variable simultaneously strengthens two or more different
response forms.
IRRELEVANT
1. The specific type of controlling variable
2. The number of response forms that are strengthened
3. The type of elementary verbal relationship
All specific features of either the stimulus or the responses are irrelevant in determining
whether or not verbal behavior is multiply controlled. Furthermore, the different response
forms may be emitted at different times or at about the same time. Any complex stimulus
situation may affect more than one response. A dog walking by may simultaneously increase
the likelihood of saying “dog” and of saying “canine.” You may say both words on the same
occasion or days apart. It may also increase the likelihood of the responses “fur,” “German
Shepard,” “brown,” “walking,” and others. Of course, it is typically the case that only one of
these responses will actually be emitted, probably because several other controlling variables
are also affecting which response occurs at this time.
This is an example of multiple responses in a tact relation, but the same process also occurs in
other elementary verbal relations. For example, the auditory stimulus “dog” may increase the
likelihood of the echoic response “dog” or the intraverbal “cat” or the tact “common noun.”
75
4-1 STUDY FRAMES
Write MULTIPLE RESPONSE (MR), MULTIPLE CONTROLLING VARIABLES (MCV), BOTH or
SINGULAR CONTROL (SC) in the blank. Then record your answers by doing the activity for this
Study Frame Set on the CD.
1. Saying “aardvark” as a result of hearing someone say, “What is that animal in the
cage?” when an actual aardvark is in the cage. __________
2. Saying “aardvark” as a result of seeing an aardvark and saying “armadillo” as a
result of seeing an armadillo. __________
3. Saying “There is a dog, a German Shepard” as the result of seeing a German
Shepard dog. __________
4. You see a new pickup truck in your neighbor’s driveway and say “Nice truck.” The
next day you see the truck again, but say “Nice pickup.” __________
5. The response “double” in the ad: “Your Xerox repair man will be there on the
double.” __________
6. Saying “bow” (rhymes with “go”) when you see an archer write bow, but later
saying “bow” (rhymes with “cow”) when seeing an actor write bow. __________
7. In the presence of a red square, you say “red” when asked “what color?” and you
say “square” when asked “what shape?” __________
8. You see the text Describe Santa Claus. You say “describe Santa Claus--white beard,
red cheeks, fat, laughs…” __________
9. Saying “Mmmmm” as a result of smelling a pie baking, and saying “Ugggh” as a
result of smelling rotting trash. __________
10. Saying, “Please pass the salt,” as a result of seeing a salt shaker in front of your
dinner companion. __________
76
4-2 FRAGMENTARY SOURCES OF SUPPLEMENTARY STRENGTH
In all of the examples of multiple controlling variables we have examined so far, each
of the controlling variables affected the entire response form. This, however, need not always
be the case. Sometimes one variable will determine only part of the subsequent
response. When this occurs, that variable is called a fragmentary source of strength.
Definition: FRAGMENTARY SOURCE OF STRENGTH is a type of stimulus control that has
the following features:
DEFINING:
1. The response form is multiply controlled.
2. At least one of the controlling variables strengthens only part of the
response form.
IRRELEVANT
1. The types of elementary verbal relationships that are involved
2. The response topography
A common example of this phenomenon is alliteration in poetry.
“The faint fresh flame of the young year flushes from
Leaf to flower and flower to fruit.”
The theme appears to be along the lines of “new,” “beginning,” “youth,”
etc. Following “faint” the form “fresh” becomes more likely than the thematically equivalent
“new” or “beginning” due to the additional (formal) source of control from the first part of the
response “faint.” “Fresh,” then, is more likely to follow “faint” than is any of many other
response forms along the same theme. “Fresh” is the result of the multiple control from the
77
theme and from the formal control by part of the previously emitted response form
(“faint”). This is a kind of multiple control where the primary control is from one source and a
secondary source of control is from a formally similar fragment of an earlier response. A
detailed treatment of multiple control in alliteration and other literary devices is presented in
Chapter 9 of Verbal Behavior.
If the multiply determined response is completely affected by a formal source, then
the strengthening is not fragmentary. If you say “no smoking” both because it would be
currently reinforcing to you if someone nearby stopped smoking and also because you
faintly heard someone say “no smoking,” the echoic source would not be considered
fragmentary (although this would be an example of multiple control).
78
4-2 STUDY FRAMES
In each case, the underlined word is multiply controlled. The main source is thematic. Classify
the second source as either a FRAGMENTARY or NONFRAGMENTARY source. Then record your
answers by doing the activity for this Study Frame Set on the CD.
1. Saying “vinyl is final.” ________________________
2. Saying “make like a drum and beat it.” ________________________
3. She sells sea shells by the sea shore. ________________________
4. Saying “stimulus” while reading a programmed text in a frame that says, “A
response is elicited by a st______.” ________________________
5. To solve a puzzle, you must change one letter in each of a succession of four
English 3-letter words, beginning with “eye” and ending with “lid.” You
write EYE - LYE – LIE – LID ________________________
6. On a quiz show, you have been asked the name of your 6th
grade teacher. You
say “6th
grade…Jones Elementary…reading club…Mrs. Poe! Mrs. Poe was her
name.” ________________________
7. On a quiz show, you have been asked the name of your 6th
grade teacher. You
say “…Mrs. ...Perry? Pinter? Paul? Poe! Mrs. Poe was her name!”
________________________
8. Saying, “Please pass the salt, Walt,” when the man at the table with you is
named Walter. ________________________
9. Saying “Hilarious,” as a result of hearing a joke on television.
________________________
10. Saying, “The faint fresh flame,” as a result of seeing the words on a script for a
play in which you have the lead role. ________________________
79
4-3 SUPPLEMENTARY STIMULATION
Under certain circumstances, a response that has previously been acquired in the presence of
a given discriminative stimulus may not currently be evoked by the presence of that stimulus.
For example, you may have learned to say “54” in response to the stimulus “what’s 6 x 9?” but
when someone asks now, you aren’t able to say the answer. There are many factors that
temporarily weaken the effectiveness of a given stimulus in increasing the probability of a
given response. These include the passage of time, the presence of other interfering stimuli,
presence of a stimulus that has only a few features in common with the set of stimuli under
which the response was acquired, illness, and the effects of drugs.
When a response is not forthcoming in circumstances where it ordinarily occurs, this does not
mean that its probability has not been increased to some extent by the presented stimulus.
We can often demonstrate this by presenting an additional controlling variable that
supplements the control already in effect and results in occurrence of the response. The
additional controlling variable is called supplementary stimulation. Supplementary stimulation
is a special kind of multiple control that the verbal community makes use of to evoke verbal
responses that already have some probability of occurring.
Definition: SUPPLEMENTARY STIMULATION is a type of controlling variable that has the
following features:
DEFINING
1. It is an added stimulus whose effect sums with another controlling variable to
evoke a response that already has some probability of occurring.
2. The supplementary variable, by itself, is never likely to be sufficient to evoke the
response.
IRRELEVANT
1. The type of controlling variable
This situation often occurs when we are trying to remember some event or the name of
something. Often, if provided the additional stimulation of the first sound of the word, we can
then emit the reinforceable response. Similarly some weak intraverbal supplementation
sometimes will be sufficient. If we are in a zoo and see an aardvark, and we have previously
learned the name but cannot remember it currently, someone may say, ”The name begins
with an ‘a’.” That may be sufficient to evoke the response “aardvark.” Notice that this would
also be called a fragmentary echoic source of strength. If we did not know the name of the
animal, the letter or sound “a” would not be sufficient to evoke the response “aardvark.”
80
Similarly, if we had the response in our repertoire, but there was no aardvark present, we
would have almost no tendency to say “aardvark” only as a result of hearing the “a” sound.
The supplementary source of control need not be formal. If you are looking at a python, but
can’t recall the name of this kind of snake, someone may say “Monty” and that may be
sufficient to sum with the nonverbal stimulus to evoke the response “python.” If there was
not an already fairly strong tendency to say “python,” the stimulus “Monty” could just as well
evoke other responses, such as “Hall,” or “that’s not his name,” or whatever.
Remember, to be supplementary stimulation, there must be multiple control over the
response and one of the controlling variables won’t by itself be able to control the response.
4-3 STUDY FRAMES
Write either SUPPLEMENTARY or NONSUPPLEMENTARY in the blank. Then record your
answers by doing the activity for this Study Frame Set on the CD.
1. A tendency to say “alligator” as the result of hearing someone say “alligator” in
the presence of an alligator at the zoo. ________________________
2. A tendency to say “crocodile” as the result of seeing a crocodile (but forgetting the
name) and then saying “crocodile” upon hearing the person next to you say that he just
bought a crockpot. ________________________
3. You have never learned that the animal that you are looking at is called a
crocodile, and you say to the person who says he just bought a crockpot, “You’ll love it.”
________________________
4. You have again forgotten (a common problem of yours) that the reptile you are
looking at is called a crocodile. Then you overhear some little kid say “crocodile” and as a
result you find yourself saying, “Crocodile.” ________________________
5. Someone asks you what your farmer uncle wants for his birthday and the answer is
“on the tip of your tongue.” The person then clears his throat and says, “I’m a bit hoarse.”
Suddenly you say, “I remember, he wanted shoes.” ________________________
6. A friend asks you, “Where was the last place you saw Hal?” To which you respond,
“I don’t know.” Then you say, “What were we doing? Oh, that’s right, we were having a drink.
It was at the bar.” ________________________
7. A friend asks you, “Where was the last place you saw Hal?” To which you respond,
“I don’t know.” Then your other friend says, “Weren’t we at the bar?” Then you say, “That’s
right, it was at the bar.” ________________________
81
8. You say, “I can’t remember Mike’s phone number,” to which your friend responds,
“It’s all evens going up.” As a result of your friend’s comment you say, “Oh yeah, 246-2468.”
________________________
9. You say, “I can’t remember Mike’s phone number,” to which your friend responds,
“Isn’t it two-four-something?” As a result of your friend’s comment you say, “Oh yeah, 246-
8246.” ________________________
10. Your teacher asks you, “What is the first prime number after 5?” You look around
the classroom trying to remember and you see your friend Sally holding up seven fingers. You
say, “Seven.” ______________________
82
4-4 PROMPT/PROBE
Supplementary stimulation often is provided for the speaker by another person. Sometimes it
is the case that this other person could identify the response that the speaker is likely to make
and sometimes he or she could not. In the latter case, the person may simply supply the
supplementary stimulation by accident (as in some of the previous examples); or the person
may supply the supplementary stimulation to see what responses might be at some strength in
the speaker at that time. If the person providing the supplementary stimulation can identify
the upcoming response, then the supplementary stimulus is called a prompt. If the provider of
the supplementary stimulation cannot identify the upcoming response, then the
supplementary stimulus is called a probe. In neither case does it matter whether the
supplementary stimulation is formal or thematic.
Definition: A PROBE is a discriminative stimulus with the following features:
DEFINING
1. It is a supplementary stimulus.
2. The person providing the supplementary stimulus cannot identify the response
that the speaker is likely to emit.
IRRELEVANT
1. Whether the controlling relationship is formal or thematic
2. Who provides the supplementary stimulus (either another person or the speaker
himself)
Definition: A PROMPT is a discriminative stimulus with the following features:
DEFINING
1. It is a supplementary stimulus.
2. The person providing the prompt can identify the response that the speaker is
likely to emit.
IRRELEVANT
Same as PROBE
83
Whether or not the person providing the prompt or probe can identify the likely response
depends on whether his own behavior is likely to be controlled by the controlling variable to
which the supplementary stimulation is added. Two people may both be looking at an
aardvark and one of them cannot remember the name. So the other says, “It begins with an
‘a’.” That is an example of a prompt because the provider is also stimulated by the primary
variable (the animal) and can emit the response.
A person may say that she saw an animal at the zoo today but cannot remember the name of
it. If you ask, “Did its name begin with an ‘a’?” that is a probe. You are the person providing
the supplement and you are not being affected by the animal that was seen by the speaker,
which may have been an alligator, or an antelope, or an aardvark – or an elephant or a
champanzee. The provider cannot identify the upcoming response.
Prompts and probes may be either formal or thematic. If the supplementary stimulation has
partial point-to-point correspondence with the response, then it is a formal prompt or probe.
For example, if the supplementary stimulation is echoic or textual, then the prompt (or the
probe) is formal control. If there is no point-to-point correspondence (that is, the
supplementary stimulus is nonverbal, intraverbal, or audience), then the prompt or probe is
thematic. A final point is that the speaker can supply himself or herself with the prompt or
probe; it needn’t be another person. Notice that this makes the prompt situation more likely,
since often the speaker will be affected by the primary source.
4-4 STUDY FRAMES
Classify each of the following supplements as PROBE, PROMPT, or NEITHER and as either
FORMAL or THEMATIC. Then record your answers by doing the activity for this Study Frame Set
on the CD.
1. A person is trying to remember another name for a pig. Expecting him to say
“hog,” you say, “It rhymes with dog.”
a. __________________________________ b. _______________________________
2. A person is trying to remember another name for a pig. You say, “Have you had
your swine flu shot yet?”
a.__________________________________ b. ________________________________
3. You ask a person standing on top of a lookout tower what he sees.
a. _________________________________ b. _________________________________
4. A Rorschach ink blot test.
84
a. _________________________________ b. _________________________________
5. You record a series of vowel sounds and then play them back, telling someone it is
a message of some kind and asking her to tell you what is says.
a. _________________________________ b. _________________________________
6. Someone is trying to identify a photograph of a person. You say, “That’s Aunt
Gertrude.” The other person then says, “You’re right, it is Gertrude.”
a. _________________________________ b. _________________________________
7. While doing a crossword in class, you pass a note to your neighbor that reads,
“What’s another word for dog tooth?” On the returned note you read the response, “K-9” You
then write “canine” in your puzzle.
a. _________________________________ b. _________________________________
85