ABA or Vebal Behavior?

fyang

有个问题想请教大家，Vebal Behavior 和ABA 哪个更适用于中等程度的自闭儿呢？ 小儿的理解力不算太差，在国内南宁方舟的时候在高级班跟不上，在中度班属于比较好的学生，到 美国后也还适应。好不容易争取到了home training,结果是verbal behavior.进行了三个月，效果不理想,进度很慢。本人感觉是因为教的太散，随机从goal里问些问题，需要两秒反应，两次section 为 90% 算通过。两小时内，一般会重复这些问题两遍，而且也很随机。其他时间基本是play time.小儿经常是玩疯了然后对问题没反应。不知道有没有家长了解Verbal behavior？特别是在美国的家长。 是Verbal Behavior 还是ABA 更适用中等程度的孩子？奇怪很多BCBA 都喜欢号称用Verbal behavior。

这段时间非常苦恼，也不知是小儿程度太低还是program的不适问题？ 请多指教。谢谢！[EM12]

秋爸爸

VB是ABA中的一个变种而已，本质上一回事儿，VB自认为是洛瓦斯模式ABA的提高。
楼主国内用中文，美国用英文吧，这个因素有考虑吗？包括家人和训练师之间的交流。还有home trainning的模式与去机构的模式之间应该有差异。
个人观点。

fyang

谢谢秋爸爸的回复。我感觉VB还是不同于ABA。小儿在ABA 中学的还可以，但一到 home training 就不行了，不是不反应，就是反应慢 (2 秒反应)。学校的ABBLS 评估也比 home training 的ABBLS高比较多。可能环境和老师变了，再加上VB讲究 natural的学习，很随机的提问和学习，这可能都是原因吧。学校用的是ABA，小儿目前在公立学校的特教中心，下半年上二年级了，每天有1-2小时的融合。双语确实是问题，不过来了一年半也还能适应。当初移民也是冲着能上学校到22岁，以后能做个简单工作应该没问题。再观察一段时间吧，否则就要求换回ABA。

秋爸爸

个人觉没必要换吧，（要换就能换？美国厉害！），充分和训练师交流一下，答疑解惑，如ABBLS的得分不一致的问题。
国内没见到用VB的，ABBLS倒是有用的。我印象中VB是比洛式ABA要“软”些，但恐怕不能认定你儿子吃硬不吃软。自然教学就是难点。
真希望楼主常来，可以开个宝贝进步专栏，介绍传授些VB的方法和见闻，捧！

fyang

秋爸爸还是不了解美国情况。依法律，公立学校是必须要给每个孩子合适的教育的，我可以拿出证据说明VB 不适合我儿就行了。对学校来说，钱付给谁都一样的，反正BCBA 都是请来的。而且这些BCBA们不大听家长意见的，认为家长不懂。VB 对训练师的要求很高。其实最担心的是这种自然教育是否适合自闭儿，在美国大部份学校也还是ABA。 真是愁死了，还得再预约些专家咨询咨询。

niuniuma

不一定是VB的问题，可能是训练师质量的问题，听起来不太负责，打发时间？---“其他时间基本是play time.小儿经常是玩疯了然后对问题没反应。”

LittleKoala

老枪的VB用的应该是很有经验，很想听他侃一侃。不过前两天他在另一个栏目刚起了个头就又不见了。

fyang

所以是很发愁，自己不了解VB，也没个监督机制，就是不负责也不知道怎么说。举个例子，关于身体部位的，小儿掌握的已基本符合ABLLS规定了，结果还要教“"Bottom",这么抽象的词，我都不知道，何况作为英语是第二语言的自闭儿。希望老枪能上来讲讲VB到底是怎么个回事。

fyang

忘了还有件事，感觉他们好象要拉长战线似的，BCBA 已经念叨了好几次说要训练到22岁，我想学校给几年home training 应该是取决于ABBLS的进程，如果都差不多了，估计学校就有权不给了。所以。。。唉，如果真是那样，即耽误了小儿时间，又浪费了学校的钱。

老枪

来的晚，估计楼主的问题解决得差不多了，空洞地说几句。

VB比LA（Lovaas 模式的ABA）出来的晚，至今也就10年，因为也是ABA，很自然地会号称VB有的东西LA没有，而LA有的VB全有，而且还比它做的好。RDI就不敢这样讲，有行为问题，不服从指令，啊啊，ABA回炉去，准备好了再来。

但这只是说说而已。就像打乒乓球，直板只能单面攻，横板能两面打，虽然刘国梁搞出个直拍横打，那个纯属花枪，没见他在重要比赛里用过。所以，理论上横板比直板优越，可是你要真碰到一个业余体校呆过的拿直拍的，别说横板，你拿门板都不一定打得过。重要的不是哪个门派，而是那个具体的consultant 或者 therapist 的实际操作能力。以前有个一个怎样找BCBA的帖子，好像有很多好的建议。另外这个论坛上美国的家长很多，有的孩子和LZ家的年龄相仿（6--9岁？），在对付学校及consultant 方面很有心得，不妨多问问。我自己的孩子小一点，不到4岁，所谓的经验只限于小小孩，而且还是自家的这个“个案”。

关于VB，按GRGD，它的流程一般是这样：开始之前，老师先要弄清楚孩子会什么，按tact,mand,introverbal,echoic,receptive 等分类，分别做成卡片，比如:
1. 中国的首都是哪儿？（introverbal）
2， 这是什么（展示陕西华南虎，tact），谁拍的（introverbal）
3. 跟我学，吃葡萄不吐葡萄皮 （echoic）
4. 站起来，单脚跳3下（receptive）
然后确定要教的孩子不会的内容，比如：
A 在黑马，白马，驴，花马图片中，找出不是马的 （receptive C56）
当然，也要选好各种级别的强化物。至于为什么要教这些东西，都是来自ABLLS，自有它的道理。

接下来正式开始一个Session，
（1）pairing：也就是跟小孩套近乎，瞎玩，或者有目的按NET教学。5-15分钟不等。很重要的一个过程，别心疼钱和时间。
（2）测试（probe）：把卡片A拿出来，摆上四张图，如果孩子挑不出驴来，写上No。
   VB记录数据和LA是有些区别的。一个老师对一个小孩只测试一次，一个问题没有任何辅助情况下能较快地答出，yes，其他所有情况，No。然后这个Probe data的表格收起，不再用。在随后的教学里，即使孩子在辅助或没有辅助的情况下会了，也不做记录。如果孩子连续3天的测试时间里能找出驴来，这个项目算掌握。如果对了两天，第3天错了，复零，重新记数。这里没有一个百分比的概念，所以楼主讲的90%，让我很不解，也许你们那里做了某些矫正。
（3）桌面教学
   1234A混合，洗牌，一张一张的问，会的要答的快，否则归到未掌握内容；问到A时，直接告诉哪个是驴，当然要讲解一下驴和马的特点，等等。下一轮时，1234换成其他会的内容，A不变。......
(4)NET教学
   离开桌面，和小孩玩。一般会选择某个具体的游戏，游戏过程中发展相关的mand，introverbal等技能。老师应该会有一个大致的NET计划。这里提到的术语，在我的另外一个“有用吗？”帖子中那个EO的链接里都有解释，而且那里还有很多记录数据的表格。
（5）桌面教学
（6）Net
....

我想对6-9岁的小孩，桌面的教学应该多一点，因为这个毕竟最接近一般的学业传授形式。如果孩子能坐住，注意力能维持，就应该多教一点，堆在桌上的卡片应该很厚很厚。但要教的内容（孩子没有完全掌握的），都会事先写在一个叫Program sheet的表格上，所以应该不是那么随机。教bottom这个词没什么不好，至少可以检验孩子学新词的能力。至于ABA到22岁，如果是说哲学，再来个22也没事儿。如果是1对1地ABA到22岁，啊啊，人家是BCBA，姑妄言之，姑且听之，由她去吧。

下面是段相关视频，因为是比较初级的水平，并没有用提示卡。注意老师要教的是elbow，之所以和他会的内容混在一起，一来锻炼学生的反应能力，其次保持住学生的注意力，另外可以形成良好的节奏和回答问题的惯性。
http://youtube.com/watch?v=Dnir_xPqOxQ

秋爸爸

猛顶！

秋爸爸

不知道ABLLS现在到了第几版？我手头上的是2.3版的，所以看到老枪讲的receptive language C56，去查时傻眼了，我那本最高到C52。v2.3中select non-examples对应的是C49，落伍啦。

jinglenn

我们这里还没有专门做VB的，但VB是我在学习ABA的过程中的挺重要的一块。我感觉这一块的重心在ABA里是讲，语言也是一种behavior，所以可以通过行为干予的方式来训练语言。

附一些书上的章节，下面这一本是导师重点推荐的。

Baum, W. M. (1994). Understanding Behaviorism: Science, Behavior, and Culture. NY: Harper Collins. (Chapter 7, pp102-129).

What is verbal behavior?
Verbal behavior is a type of operant behavior. It belongs to the larger category of behavior that could be called communication, except that communication suggests a mentalistic theory foreign to the behavioral point of view. As we shall see, the behavioral view would either redefine communication or replace it with other terms.

Communication
Communication occurs when the behavior of one organism generates stimuli that affect the behavior of another organism.
The conventional view holds that in communication something passes from one person to another. By its derivation, communication means “to make common.” What is made common? An idea, a message, a meaning. Some psychologists embellish this everyday conception by adding that an idea is encoded by the sender, passed in code to the receiver, and then decoded by the receiver, who then possess the message.

Verbal behavior as operant behavior
There is a crucial difference between alarm-calling and talking. The bird’s alarm call is a fixed action pattern, whereas talking is operant behavior. When a fixed action pattern generates auditory or visual stimuli that affects the behavior of others, that can be called communication. However, it is not verbal behavior. Even the human eyebrow flash of greeting, although it affects the person who see it and therefore is communicative, is not an example of verbal behavior.
Communication is the larger category. All verbal behavior could be called communication, but the reverse is not true. Fixed action patters depend only on antecedents (sign stimuli), whereas verbal behavior, because it is a type of operant behavior, depends on its consequences.
Verbal behavior, like other operant behavior, tends to occur only in the context in which it is likely to be reinforced.
Those around who hear and reinforce what a person says are members of that person’s verbal community – that group of people who speak to one another and reinforce one another’s utterances.

Skinner defined verbal behavior as operant behavior that requires the presence of another person for its reinforcement. This other person, who reinforces the speaker’s verbal, is listener.

Most verbal behavior depends on social and conditional reinforcement. When you and I are in conversation, we take turns as speaker and listener, my verbal acts serving to reinforce your verbal act and vice versa.

Like other operant behavior, verbal behavior requires only intermittent reinforcement to be maintained. Generally, verbal behavior is highly persistent and often reinforced only intermittently.

As with other operant behavior, verbal behavior requires less reinforcement to be maintained than to be acquired. Like much other operant behavior, verbal behavior is shaped over time by successive approximation.

Human children appear to be so constructed as to be likely to imitate speech sounds they hear from significant others. Between that genetically programmed predisposition and the reinforcement provided by these significant others as listeners, verbal behavior is acquired and shaped.
We learn to respond to the heard utterances of others as verbal discriminative stimuli. We discriminate between vocalizations and noises and between one vocalization and another. By the age of 18 months, a child normally behaves differently to “would you like a cookie?” and “Would you like some juice?”
Our actions as listeners provide reinforcement for the utterances of speakers around us. We frequently do this unconsciously. Along with speaking, listening is frequently and lavishly reinforced in small children. When the mothers responded to the infants’ vocalizations, the mothers were reinforcing both speaker-behavior and listener-behavior because the children spoke in the context of hearing whatever their mothers had just said.
As time goes on, differential reinforcement refines the child’s listening. A parent says “pick up the red ball,” and when the child picks up the red ball rather than one of another color, delight, praise, and affection follow. Thus our listener-behavior is reinforced and shaped. The listener’s actions bust be reinforced to be maintained just as much as the speaker’s actions.

jinglenn

Examples
The importance of history. Suppose a stranger begins speaking to you in Russian, and you don’t understand a word. There is no possibility of this behavior being reinforced. Is this verbal behavior? Are you a listener?
Even though the stranger’s speaking cannot be reinforced in this situation, it can be called verbal behavior because it was reinforced in past situation by the stranger’s verbal community. That it cannot be reinforced on this occasion in no way disqualifies it, because verbal behavior often goes unreinforced on particular occasions. This sort of behavior qualifies as verbal because it arises from a history of reinforcement by a community of speakers and listener.
The answer to the question of whether you are a listener even though you cannot understand a word of Russian depends not only on a history of reinforcement, but also on perspective. From your perspective, you cannot be a listener because you cannot reinforce the stranger’s behavior. From her perspective, however, you are being treated as a member of the stimulus class “listener”. She will soon discover her mistake and discriminate – that is, she will go elsewhere or speak to you in another language. Her speaking to you in Russian can be thought of as an instance of generalization. As a discriminative stimulus, you look enough like a Russian listener for her behavior to occur. Your inability to reinforce the behavior insures that it will extinguish in your presence, but the initial action arose from a history of reinforcement in the presence of listeners a lot like you. From the perspective of that history, you are initially a listener, or at least a potential listener.

Sign language and Gestures. Suppose you and the stranger share no common language, and she resorts to gestures. She points to her wrist and looks at you questioningly. You show her your wristwatch, and she nods and smiles. Do her gestures count as verbal behavior?

According to our definition, they do. Her pointing to her wrist is operant behavior, the reinforcement of which depends on your presence.

According to our definition, verbal behavior need not be vocal behavior, and it can even be written. (A verbal behavior is an operant behavior that requires the presence of another person (the listener) for its reinforcement.

The best example of nonvocal verbal behavior is sign language. The silent singer acts as speaker, and the one who responds to the signs, though deaf, is the listener. A group of singers who alternately play speaker and listener roles constitute a verbal community.

Nonhuman Animals. My cat comes to me at dinner time, meows and rubs against my leg. He does this every day, and every day I feed him when he does this. Is my cat’s meowing verbal behavior?

According to our definition, it might be. The meowing is operant behavior because it arose from a history of my reinforcing it by giving him food. It requires my presence for it to be reinforced. That would make me the listener and my cat the speaker.

However, you might disqualify my cat’s meowing as verbal behavior because my cat and I cannot reasonably be called a verbal community. We never exchange our roles of speaker and listener. I never ask him for food, nor does he ever feed me. He sometimes comes when I call, but that seems to flimsy a reason to call us a verbal community.

Yet this example makes a point: the definition of verbal behavior in no way excludes nonhuman animals. Chimpanzees have been taught to communicate with humans by means of sign language. Although my cat and I may fail to qualify as a verbal community, when a chimp and a human sign back and forth, they do qualify. Just as two humans signing back and forth alternate as speaker and listener, so do chimp and human. Instances have been reported in which two trained chimps signed back and forth to each other. According to definition, this might qualify as verbal behavior if the two chimps could reasonably be considered members of a verbal community.
Many thinkers have argued that language is uniquely human. Whether this is true depends entirely on the definition of language. If it is defined in terms of speech so as t exclude gestures, then of course it belongs only to humans. The definition of verbal behavior could be similarly narrowed so as to exclude nonhuman animals. Such definitions, however, would deny language and verbal behavior to signers. The present definition, by requiring that speaker and listener be able to exchange roles, rules out trivial cases like my cat and me, but because it includes gestures, allows the possibility of verbal behavior in nonhumans.

Talking to myself. When I talk to myself, is that verbal behavior? Behavior analysts disagree about this. Their answers depend on whether they accept the idea that the speaker and listener in a verbal episode can be the same person.

If the same person can be listener and speaker, then the verbal behavior of the speaker is reinforced by a change in behavior on the part of the same person (viewed as listener). This may happen, for example, when I instruct or command myself. In driving to an unfamiliar house, I might say to myself at an intersection, Now, here I should turn left. If as a listener then turn left, that reinforces the verbal act (the self instruction to turn left), particularly if I successfully reach my destination.

Whether covert or overt, the taking to oneself would have to result in change in one’s behavior (as listener) before it could be called verbal behavior. Self-instructions and self-commands qualify. Even a declaration, such as That is a beautiful painting, would qualify if the one then took some action like looking up the artist’s name or asking the price.

Behavior analysts who reject the idea that talking to oneself constitutes verbal behavior regard it as part of an extended unit of action. Such extended or molar unites play a large role in Rachlin’s approach to operant behavior. In this view, my driving to the unfamiliar house would constitute a unitary act, a functional category. The at might occur in a variety of ways, following different routs , with or without self-instruction, but all the variations could be considered members of the same functional category. If driving with self-instruction gets me to my destination more often, then I will do that more often. Most likely, I will drive that way the first few times I drive to that particular place, and as the route becomes familiar, driving without self-instruction will take over. When behavior is thought of in these molar terms, an act cannot qualify as verbal unless the listener differs from the speaker.

jinglenn

Verbal behavior versus language

Verbal behavior differs from language. The word language, when used in phrases like “the English language” or “American sign language,” seems to be a thing. Language is often spoken of as a possession, something that is acquired and then used. The common idea that language is used like a tool raises all the problems of mentalism. Where is this tool? What is it made of? Who uses it, how, and where? How does this tool cause speech?

Verbal behavior comprises concrete events, whereas language is an abstraction. The English language, as a set of words and grammatical rules for combining them, is a rough description of verbal behavior. It summarizes the way a lot of people talk. It is rough, because people often use poor English. Neither the explanations in a dictionary nor the rules in a book of grammar exactly coincide with the utterances of English speakers.

Although talk of using language is mentalistic and misleading, when we say a person is doing this, that person is usually engaging in verbal behavior. Instances of using language that might not be considered verbal behavior could be events like writing a book or reciting a poem to oneself. Conversely, some instances of verbal behavior, such as waving and pointing, might not be considered using language.

Functional units and stimulus control

Like other operant behavior, verbal behavior consists of acts that belong to operant classes that are (1) defined functionally and (2) subject to stimulus control. These two ideas set the concept of verbal behavior apart from traditional view of language and speech.

Verbal operants as functional units

Every event can be said to have a structure, and probably each event has a unique structure. A rat probably cannot press a lever twice in exactly the same way using exactly the same muscles to exactly the same extent. In contrast, functional units are not particular events, but classes of particular events, and are defined by their effects in the environment.

A verbal operant is a class of acts, all of which have the same effect on the listener. For example, all the structurally different ways of requesting the salt belong to the same operant because they all have the same effect on the listener – getting the salt passed. Function can be understood only from circumstances and effects.

Stimulus control of verbal behavior

As with other operants, a verbal one becomes more or less likely to occur depending on circumstances – that is, depending on discriminative stimuli. One reason that a ward cannot constitute a functional unit is that the same word can serve different functions, depending on the circumstance. For example, the word “water” may have different effects on the listener under different circumstances, so in each context water would belong to a different verbal operant.
The relation of circumstances to the likelihood of the verbal operant is the relation of stimulus control. No strict one-to-one correspondence exists between a discriminative stimulus and a verbal operant, the way there might be between a tap on the knee and a jerk of the let. The discriminative stimuli only modulate and make it likely that instances of certain verbal operants will occur.

Among the most important discriminative stimuli modulating verbal operants are auditory and visual stimuli generated by another person acting as speaker. Having played listener fro the other person, I may play the role of speaker and generate discriminative stimuli that affect the other person’s behavior.
All operants occur within a context, and the enabling effect that the context has on the operant arises from the history of reinforcement associated with that context. As with other operants, so with verbal operants.

Like other operants, verbal operants cannot be defined solely in terms of their consequences. The context usually needs to be specified also. Requesting directions from a stranger differs from requesting directions from a familiar person. If you were studying politeness theory, which concerns the way that verbal behavior depends on the person addressed, you might want to make the distinction. For other purposes, “requesting directions” or even “making request” might be fine enough.
Common misunderstandings

The idea of verbal behavior emphasizes the similarity of speaking and gesturing to other types of operant behavior. Conventional views try to set language related behavior apart, to define it as special and different. Three characteristics that have been urged as unique to language-related behavior are (1) that it is generative – that people constantly generate novel utterances; (2) that speaking, unlike other behavior, can refer to itself; and (3) that speaking, unlike other behavior, can refer to events in the future. Let us see if these really set language-related behavior apart from other behavior.
The Generative nature of language. Every day you generate utterances you have never made before. Probably most sentences you speak are novel in this sense. In fact, each utterance is unique, because you cannot make two exactly the same. Non-repeatability characterizes all operant behavior. It is not that verbal behavior is varied and other behavior is fixed, as the conventional view would hold, but that all operant behavior is just as varied as verbal behavior.
Critics of this view point to the importance of grammar in generating utterances. Grammar is a part of any language, and grammatical structure is that is offers a rough description of the structure of some verbal behavior. Real speech is frequently and sometimes mostly ungrammatical. Our sentences often break the rules of sentence construction, and we often leave sentences unfinished.

Still, spoken English generally follows the order subject-verb-object, and there are other such regularities. But the rough structural regularities that characterize verbal behavior also characterize other behavior. A regular sequence of motions goes into each of a rat’s lever-presses. Each lever-press could be equated with a sentence, and we could write a grammar of lever-presses. Only certain sequences of motions result in the lever being pressed; those would be the permissible “sentences”. As before, the characteristic that is supposed to set verbal behavior apart can be seen as shared by other behavior.
Talking about talking. Linguists and logicians make much of statements called meta-statements, which refer to themselves or other statements. Meta-statements form the basis of some arguments that the ability of language to refer to itself sets it apart from other behavior. Viewed as an utterance – as verbal behavior- however, there is nothing magical about it. It conforms to the standard English sentence frame of subject-verb-attribute. The only unusual aspect of this particular utterance is that the subject is an utterance.

To the behavior analyst, meta-statements are talk about talking – that is, verbal behavior under stimulus control of other verbal behavior. Talk about talking occurs all the time. If you didn’t hear what I said, you ask me what I just said and I repeat it. Your question plus what I said, which I heard perfectly well, constitute the discriminative stimulus for my repeat utterance. My ability to do this derives from a long history of reinforcement for this sort of repetition’ we are trained from an early age to repeat utterances for effect and on cue.

Reporting an utterance on request is an example of a verbal self-report, which is verbal behavior partly under the control of one’s own behavior as discriminative stimulus (either verbal or nonverbal). We can report also on private verbal behavior, as when you ask what I am thinking, and I say I was thinking how nice it would be to go to the beach today. My verbal behavior then is partly under control of a private discriminative stimulus.
Talking about the future.  When you talk about verbal behavior you are inclined to engage in, it sounds as though you are talking about the future. Since future events cannot affect present behavior, people are tempted to invent a cause in the present – an inner purpose or meaning - and even to insist that talk about the future proves the existence of mental images.

Nothing need be going on in my mind or anywhere inside for me to make utterances that seem to refer to events in the future or, for that mater, any other events that have never occurred. I have never seen a purple cow, but I have pronounced purple and cow and put adjectives together with nouns. My utterance including the phrase purple cow in no way requires that I have a purple cow in my mind or anywhere else. It only requires that I have a history of reinforcement for the sort of verbal behavior that people often call imaginative.

Similarly, if I talk to you on Monday about an appointment we will have on Friday, I need have no ghostly image or meaning in mind. Making and keeping appointments is operant behavior arising from a long history of reinforcement. You tell me you want to see me. Hearing this auditory discriminative stimulus, I write in my appointment book and say I will see you on Friday at 3:00. doing and saying this sort of thing has been reinforced many times in the past.

jinglenn

Meaning
In the conventional view of language-related behavior, words and sentences have meaning, and the meaning contained in an utterance is passed from speaker to listener. To a linguist interested in a formal analysis of the structure of English, such a view might do little harm. As a theory of the behavior of speaking or verbal behavior, however, it suffers all the sortcomings of any mentalistic theory.

Reference theories
Philosophers and psychologists, trying to turn the rough everyday notion of meaning into a more definite theory of language, invented theories that rely on the notion of reference. The word dog, for example, whether spoken, written, or heard, is said to refer to the sort of four-lagged mammal that arks.
Symbols and lexicons. The notion of reference suggests that the different forms of the word dog – spoken, heard, written, seen – are symbols for the category of actual dogs. How can all these symbols are recognized as equivalent? All the different symbols are somehow connected to something inside. Since actual dogs cannot be inside the person, some representation of the category is supposed to exist somewhere inside, and all the symbols for dog are said to be linked to this representation.

Where is this representation? It is said to be in a lexicon, a collection of such representations of objects and events of the real world. The speaker is said to look for the representation in the lexicon, find it there connected to its symbols, and then use the appropriate symbol. The listener is said to hear the symbol, look up the symbol in the lexicon, find it connected with its representation, and then understand it.

The idea of reference was probably invented to explain equivalences. Add to this variety the words for dog in different languages, and you may see how tempting it is to suppose that all these acts and stimuli are equivalent because they are all somehow tied to some representation or meaning somewhere inside.

It is altogether too easy to suppose that the observed equivalence arises from some ghostly inner equivalence. But where did the observed (or ghostly) equivalence come from. We come to do this over time, after exposure to these different stimuli, and after a history of reinforcement for the appropriate response.

The importance of context. Not only do reference theories offer no account of speaking, they fail even at the task for which they were invented – making sense of meaning – because they cannot take account of context. (e.g., water means differently in different situations, a request, a question, the naming of a liquid on the floor, and the naming of an ingredient.) If context determines the meaning of concrete nouns like water, how much more fundamental it is to the meaning of abstract nouns and utterances composed of many words (e.g., most people in the United States regard poison ivy as a weed; whereas people in some Scandinavian countries consider it an attractive plant for landscaping. The word weed depends as much on the circumstances as it does on the plant.) Reference theories have an even worse time with actual utterances containing several words. Suppose my son and I are building a brick wall. My job is laying the bricks, and his job is handing me bricks. Again and again, I ask for a brick. I say, Hand me a brick, Let’s have a brick, Brick!, I need a brick, Give us a brick, and many other variations. Sometimes I just turn and look or hold out my hand. All of these acts have the same meaning. You could not find the meaning by looking in a lexicon because their meaning lies in what the acts accomplish: getting my son to hand me a brick so that we can proceed with the wall.

Meaning as use
Instead of talking about meaning, behaviorists talk about the use or function of an act or utterance. Roughly speaking, that is the meaning of meaning.
Consequence and Context: Suppose I put a rat in a chamber with a lever and a chain. Pulling on the chain produces food; pressing on the lever produces water. The rat pulls and eats, presses and drinks. You could say that the meaning of a chain-pull is food and the meaning of a lever-press is water. A person in the same situation might make the sound food and receive food, make the sound water and receive water. Are these situations fundamentally different? The behaviorist says no. the rat has had no food for a time – it pulls the chain and gets food. John has had no food for a time – he says food and gets food. Either way, the use of the act consists of its consequences in the context.

The meaning of verbal behavior is its use its consequences in the context. The context in which someone would emit an utterance including the word meaning tells us the meaning of meaning.

As with other operant behavior, verbal behavior depends on a history of reinforcement. To say that the use or meaning of a verbal operant is its consequences in the context is to say that is occurrence depends on a history of such consequences in such contexts. My children learned to say please when they made a request because again and again reinforcement was available only for a request including that word.

Mands and Tacts. In every terms, verbal operants serve a variety of purposes. Two of the most important are to request and to inform. Mands include not only request, but commands (e.g., the army sergeant who says left face!), questions (e.g., My asking you what time is it?) and even advice (e.g., a parent tells a child you should take algebra this year). The exact setting in which the request, question, or advice may be emitted can vary widely, and yet we still recognize it as the same mand – whether Bob asks Jane for salt or whether he asks Tom, Dick or Harry – because the reinforcer is the same. When the reinforcer of a verbal operant is well-specified, the operant is a mand.

Verbal operant that might be considered informative specify no particular reinforcer; rather occur in the presence of some particular discriminative stimulus. The whole point of the utterance there is a tiger behaved you is the tiger; the reinforcer that the listener will provide remains unspecified. Skiner called such verbal operant tacts.

Tacts include a wide variey of utterances. Opinions and observatios are tacts. replies to questions are often tacts: you look at your watch and tell me the correct time. What we have been calling verbal reports are all examples of tacts.
Dictionary definitions. If words and utterances cannot be understood by their inherent meaning, then why should we bother with dictionary definitions? Let us rephrase the question. How are dictionary definitions helpful? When I come across an unfamiliar word an consult a dictionary, I do not learn the meaning of the word; I get a summary of how the word is used, usually with one or more examples and some synonyms (different words that might occur in similar circumstances or have similar effects) and antonyms (different words that occur in contrasting circumstances or have contrasting effect). All of this helps to guide my behavior as reader, listener, speaker and writer.

Thus, dictionaries do not contain meanings. They exemplify the general way that we learn how to use words, by hearing and seeing them used. How did you learn jump, run, talk, car, and baby? Most of the words we use we never look up in a dictionary, and no one ever defines them for us. If this were not so, dictionaries would be useless, because they explain how to use a word in terms of other words that are supposed to be familiar already.
Technical terms. What is true of everyday words you might look up in a dictionary is doubly true of technical terms invented by scientists and other professionals. A term is aways defined in terms of others. Sometimes a set of interrelated terms are equally familiar (or unfamiliar) yet are still all defined in terms of one another. Consider the term trait, gene, and inherit. None can be defined without using the other two. So, too, with the terms of behavior analysis: reinforcement, operant, discriminative stimulus. What is operant behavior? Behavior that is more likely in the presence of a discriminative stimulus because of a history of reinforcement in the presence of that stimulus.

This interdependence of definitions only seems to be a problem if we insist that each term must have its own separate meaning, suitable for storage in ghostly lexicon. It poses no real problem for scientists; it is simply a feature of scientific vocabularies. Interdependence of terms just means that they tend to be used together.

jinglenn

其它的一些有关VB的笔记

Language and language-use traditionally studied by Cognitive and Developmental Psychologists. (Also a distinct area called Psycholinguistics). These approaches are unashamedly mentalistic.
There two classes of theories, Biological theories and cognitive theories.
Biological theories states that language acquisition facilitated by innate brain structures (Chomsky, Pinker). The common terms they use include: innate language acquisition device, language synthesizing, deep and surface structure, transformational grammar.
Cognitive theories states that language is the result of information-processing systems (Piaget, Brown, Pinker). The common terms they use include: encoding, referent, lexicon, storage, retrieval, decoding, expressive language, receptive language, central executive

Behavioural analyses of language-use generally follow Skinner’s theoretical analysis. Skinner published Verbal Behavior in 1957, which provided a Radical Behaviourist analysis of language based on same principles used to analyze all other learned behaviour. But the analysis was only conjectured based on his casual observations of language in every-day life. Skinner (1957) described his book as “an exercise in interpretation rather than a quantitative extrapolation of rigorous experimental results”.

The Analysis of Verbal Behavior journal started in 1982, published by the Association for Behavior Analysis (USA). Consequently, empirical research taking a behavioural perspective on language still in its infancy.

Michael (1984) discusses possible reasons for the paucity of behaviour-analytic research into language and encourages further activity. One possibility is that researchers were dissuaded by Chomsky’s (1959) harsh and bitter review of Skinner’s Verbal Behavior. Unfortunately, Chomsky’s review was never replied to by Skinner.  It seemed like Cognitive Psychologists had won the battle. 11 years later MacCorquodale (1970) described how Chomsky (1959) missed the essential points of Skinner’s Verbal Behavior and of Radical Behaviourism generally.

The behavioural literature has increased immensely in past 15 years largely due to surge in research investigating language training in autistic children.  (Autism characterized by language deficits & the behavioural language interventions found much more effective than traditional approaches)

jinglenn

Basic Principles
The term language implies some thing a person acquires and uses. Similarly, communication literally means to make common.
But what is made common? Ideas?  Meanings? The suggestion of mental operations is clear.

Woodworth (1921) made a GENERAL point regarding the construction of mental phenomena
        Instead of "memory", we should say "remembering"; instead of "thought" we should say "thinking"; instead of "sensation" we should say "seeing, hearing, " etc.  But, like other learned branches, psychology is prone to transform its verbs into nouns.  Then what happens?  We forget that our nouns are merely substitutes for verbs, and go hunting for the things denoted by the nouns: but there are no such things, there are only the activities that we started with, seeing, remembering, and so on.. It is a safe rule, then, on encountering any menacing psychological noun, to strip off its linguistic mask and see what manner of activity lies behind.

Skinner (1957) introduced the term verbal behaviour (VB) to redirect attention to environmental determinants of language use & the learning involved

Basic premise: Instances of verbal behavior are instances of operant behaviour (i.e., behaviour learned & maintained by contingencies of reinforcement). Thus, Instances of verbal behavior can be analyzed in terms of their functions and organized in terms of operant classes.  
It has TWO implications:
1.        Different verbal response forms (topographies) are functionally similar if they share antecedents & consequences (i.e., we define function by identifying SDs and SRs).  
2.        The same verbal response form can serve different functions (i.e., appear in a number of different operant classes).  

Offers a new approach to meaning
Meanings and referents are not to be found in words but in the circumstances under which words are used by speakers and understood by listeners. (Skinner, 1974, p. 103)
        The meaning of an expression is different for speaker and listener; the meaning for the speaker must be sought in the circumstances under which he emits a verbal response and for the listener in the response he makes to a verbal stimulus. (Skinner, 1974, p. 191)
        In the terminology of meaning, we say that the word doll is used at one time “to ask for a doll” and at another “to describe or refer to a doll.”  When the response Doll! has been acquired as a mand, however, we do not expect that the child then spontaneously possesses a corresponding tact of similar form.  If we find both types of operants in the repertoire of the child, we must account for them separately.  This appears to make the task of explaining verbal behavior more difficult, but the advantage which appears to be gained by the traditional concept of the “word doll” is offset by the problem which remains of explaining how a child may learn to use a word both to “express a desire” and also to “describe an object”.  The total formulation has not been simplified; part of the task has merely been postponed.  If we are to accept the full responsibility of giving an account of verbal behavior, we must face the fact that the mand “doll” and the tact “doll” involve separate functional relations which can be explained only by discovering all relevant variables. (Skinner, 1957, pp.187-188)

jinglenn

The Definition of Verbal Behaviour
Skinner (1957) defined VB as operant behaviour learned and maintained by reinforcement that involves the behaviour of others (cf. nonverbal operant behaviour which is reinforced by changes produced in the physical environment. E.g., opening refrigerator door provides access to food).

OR, operant behaviour that requires the presence of another person (an audience) for its reinforcement

Differential to social behaviour
The category of social behavior is wider than vocal behaviour, because spoken, gestural and written performances of a speaker/mimer/writer alter the behaviour of a listener/viewer/reader, and are maintained by these effects.

Differential to communication
The category of verbal behavior is not as wide as communication, because communication includes numerous unconditioned reflexes & fixed-action patterns (e.g., crying & smiling in infants, frowning & eyebrow flash in adults).

Verbal behavior refers to the behavior of speakers only.
A listener’s response to speaker may or may not be verbal behavior – called “receptive” language or verbally-governed behaviour.

The Verbal Community
Verbal Community: All those potential listeners who could reinforce the speaker’s behaviour and exchange roles with the speaker (i.e., speakers become listeners and vice versa)
1.        Behaviour of pets/infants/machines not verbal behavior because role exchange dubious
2.        Talking to one’s self not verbal behavior because role exchange dubious. (But, is it really?)

The contingencies that regulate verbal behavior arise from practices of people in the verbal community (Skinner, 1957). These practices refer to the customary ways that people reinforce the behaviour of a speaker.

An interaction between two organisms can be considered verbal behavior only when those organisms belong to the same Verbal Community (Baum, 1994):  

The Verbal Episode
Verbal Episode: The sequence of stimulus, response, and reinforcing events that define any interaction between two or more members of a verbal community.
The contingencies of reinforcement maintaining the speaker’s and listener’s behaviour in any verbal episode can be analysed in terms of functional response units (i.e., 3 or 4-term contingencies.) With the analysis, the different contingencies operate for speaker & listener are clear.

jinglenn

The Elementary Verbal Relations   
Skinner (1957) argued that verbal utterances (including gestures, writing text, etc) serve a number of different types of function.
•        different types of functional relations between verbal behavior and environment
•        functionally different types of verbal behavior
•        a number of classes of operant classes
•        a number of elementary verbal operants
•        a number of elementary verbal relations

In all cases, the reinforcing consequences involve the behaviour of others
Many topographically different responses can exist in one relation, and the same topography (e.g., saying “dog”) can exist in more than one relation. In addition, a verbal utterance might exist in more than one relation at any one time (i.e., verbal behavior and non-verbal behavior are often controlled by multiple contingencies)
The following list of elementary relations is incomplete (e.g., Skinner also offered the autoclitic) but abbreviated for sake of simplicity.


•        Echoic: Repeating what is heard (e.g., saying “dog” after hearing someone else saying “dog”)
•        Imitation: Copying someone’s motor movements (e.g., clapping after someone else claps)
•        Mand: Asking for reinforcers that you want  (e.g., saying “dog” to ask for a toy dog)
•        Tact: Naming/identifying objects, actions, events, etc. (e.g., saying “dog” in the presence of a dog)
•        Intraverbal: Answering questions or holding conversations where utterances controlled by preceding utterances (e.g., saying “dog” after someone else says “Lassie”)
•        Textual: Reading written words  (e.g., saying “dog” after seeing the written word “dog”)
•        Writing: Writing and spelling words when spoken to you  (e.g., writing “dog” after someone else says “dog”)
•        Receptive (Simple): Following instructions or complying with the mands of others  (e.g., touching dog when asked “touch dog”)
•        Receptive (complex): Identifying specific items when given some description of its features, functions, or classes/concepts (aka Receptive by Feature, Function & Class, RFFC)