Stacks Image 70

7. Stimulus Control

Take a look at the child in Movies 7.1. and 7.2.
Well, what do think? Can he read? In one sense he can read. He was shown a card with the word ‘Clap’ written on it and he behaved appropriately. But is this reading? His behaviour certainly wasn’t a reflex, so what was it? A starting point is to use our lay vocabulary and say it was ‘Voluntary’ behaviour. That is, Tara (his name) voluntarily engaged in the behaviour of clapping when the sign with the word ‘Clap’ was placed in front of him. Now that’s a fairly accurate description of what you saw, devoid of any interpretations as to why he did it. Or is it?

One word that gives cause for concern is the term ‘voluntary’. This word has all the hall marks of a mentalistic term. It describes his behaviour and at the same hints at the explanation for the behaviour, “It was voluntary!”. To delve into this analysis a little deeper we need to remember that every observation is an observation of an end product that emerges from previous interactions with the environment. If we look closely at Tara’s previous interaction with the environment what might we find to explain his behaviour? Look at Movie 7.2. In this clip you will see a game played with his mother. When deconstructed, the game involved three components.

Firstly, we see the written word ‘Clap’ -
Next, we see clapping -
Finally, we see Tara being given a bottle to drink -

The relation between these three terms is called the ‘
3-term contingency’, sometimes written as A:B --->C, or as SD: R —> S+ where SD stands for a stimulus in the presence of which a response (R) produces either a positive (stimulus added; S+) or a negative (stimulus taken away/subtracted; S-) consequence.

Before we elaborate on the significance of the 3-term contingency, let’s take a look at another way of addressing the term ‘voluntary behaviour’. This time, you have to conduct a little study for yourself. Your goal is to create a stimulus that produces voluntary behaviour and then we will examine how useful the term ‘voluntary’ is within a scientific analysis of that behaviour. Here is what you need for your study:

A piece of stiff card that is BLACK on one side and WHITE on the other side.

Now for your instructions: You can work either with a group of people or with a single person. Hold up the card and ask your participant to name the colour. Now put the card behind your back and bring it out again showing the other colour. Ask your participant to name the colour. Repeat this exercise about 10 times fairly fast, each time randomly showing either black or white; you don’t have to ask the colour each time, but your participant has to name it for you. After about the 10th card presentation, show the white card about 5 times in a row. Then, and here is the interesting bit, without presenting the card, say the following, “What do cows drink?”. When you do this your participant is likely to answer “Milk”.

The point of the exercise is to create an opportunity for voluntary behaviour to emerge and for you to then examine the nature of that behaviour. Undoubtedly, the response you get is a voluntary response, not a reflex. But here is the interesting bit. We anticipated the response, incorrect as it was, because we loaded the dice, so to speak. We arranged an interaction with the environment (you and the different coloured card) to increase the likelihood that a particular voluntary behaviour would emerge. The explanation for the word “Milk” occurring, then, is to be found from an analysis of the procedure you used; don’t use that procedure and you won’t get that response. In other words, the voluntary behaviour is a dependent variable, and the design of the environment is the independent variable that was responsible for the occurrence of the dependent variable.

Contrary to traditional ways of looking at behaviour, this analysis concludes that the term ‘voluntary’ is not very useful in a scientific analysis. In fact, as with all mentalistic terms, it stands in the way of further analysis because it implies that there is no need for further analysis. It might be said, for example, that the origins of the behaviour lie in the wilful action of the person. But, if a group of people all make the same mistake, then surely there’s a message staring you in the face. With so much freedom in their wilful actions, how come the same word is used by everyone? That line of analysis gets nowhere. Indeed, that kind of analysis would not have been able to design the exercise in the first instance. No, better to accept that the term ‘voluntary’ is often used because we don’t actually have an explanation for behaviour, and therefore the illusion of an explanation suffices until something better comes along.

Movie 7.3 is a slide show that we use in our teaching to illuminate the implications of this argument.

Returning to Tara for a moment, there is another issue that needs to be addressed. Clearly, Tara was familiar with the word ‘Clap’. But can we say he ‘knows the meaning’ of the word? It is tempting to fall into the vernacular again and say that he does. However, at this young age, it doesn’t make sense to analyse his understanding of the word in this way. The only accurate statement we can make is that both ‘meaning’ and ‘understanding’ are to be analysed by looking at the contingencies that transformed the word ‘Clap’ from an arbitrary scribble on paper (i.e., this is what it is from Tara’s perspective) to a stimulus controlling a pattern of behaviour. That is, the previously neutral stimulus now functions as a Discriminative Stimulus (SD) and the behaviour is said to be under Stimulus Control. When he grows older, he may find a verbal definition of the word in a dictionary, but in truth, the understanding of the word emerged from the contingencies to which he was exposed (Movie 7.4).

At first blush this sounds incredibly pedantic. But we can approach the ‘meaning’ of all stimuli that are familiar to us using this analysis. Some familiar examples are shown in Movie 7.5. In this movie we see a visual representation of the 3-term contingency. Examples for each component are listed. Notice that at B, there are a number of possible behaviours that can occur. This is one way to represent the notion of an OPERANT, where a class of behaviours share a similar consequence. It is the OPERANT that is brought under control of the Antecedent because of the consequences that result.

The 3-term contingency is central to the analysis of all behaviour because it identifies events that reliably appear before and after segments of the behavioural stream. In experimental work, when a stimulus is trained as a
Discriminative Stimulus then other findings emerge. For example, behaviour trained in the presence of one stimulus appears also in the presence of another stimulus without any other training. This can happen for a number of reasons, but the principle shown here is called Generalisation (Fig. 7.1).

here to read about pigeons learning to discriminate between different artists.

Figure 7.1

Pecking is trained to one colour, but when different colours are presented during a test, pecking occurs at different rates across these stimuli, with most pecking occurring in the presence of stimuli that closely resemble the training colour. Pecking is under discriminative control by the central colour (also called Stimulus Control) and it generalises to the other colours. If we replace the behaviour of pecking with the behaviour of ‘reading’ or ‘naming’, or any other behaviour, a similar perspective can be used because in each instance responding appropriately means emitting the same behaviour in the presence of a discriminative stimulus (e.g., Fig. 7.2).

Figure 7.2


When there is evidence of good
Stimulus Control a stimulus (i.e., very little generalisation) it goes without saying that we also have good Discrimination between stimuli. This helps get a better handle on a term like ‘concept’. Concepts have been viewed in behaviour analysis as generalisation within a class of stimuli and discrimination between classes of stimuli. Thus, for example, the concept of RED involves generalisation among red stimuli (i.e., we call them all red) and discrimination between these stimuli and others that we don’t call red. Other factors involved in concept formation are discussed in the chapter titled Stimulus Equivalence.

Differential responding

Introducing students to a scientific d
iscussion on discriminative control by stimuli is always hampered by one thing, familiarity. Thus, while one goal of the scientist is to enlighten the community with new findings and perspectives on familiar issues, old perspectives on these same issues make it particularly difficult to be persuasive about the value of a new perspective. Historical influences are difficult to offload. Added to this is the powerful influence of the current environment. All stimuli in our environment influence our behaviours in a variety of ways. Look around the room just now and note that everything in view has been designed by a person at one time or another. Your physical environment, the medium used to convey these words, all have been designed with a view to controlling you in one way or another.

But what if you are given the task of designing an environment so as to produce a particular behaviour, would you be able to do it? For example, what if you had the task of teaching a pigeon to discriminate between art works of two famous artists (
click the button on the right), or teaching a pigeon to respond appropriately to the words ‘Peck’ and ‘Turn’? Given that we know there are principles of behaviour (i.e., that behaviour is not a random event but something that complies with natural laws), your task can be reframed like this: ”How can I harness the principles of behaviour to facilitate meaningful changes in behaviour?” By phrasing the task this way, you are in fact asking how one can instil ways of behaving so that they become familiar to the organism concerned. A powerful technique we have already touched in the shaping exercise is ‘Differential Reinforcement’ (Download PPT presentation). Both of your tasks involve differential responding to different stimuli.

The videos below show examples of the outcomes arising from differential reinforcement. As you watch, keep an eye on your tendency to engage in mentalism.


‘Differential responding’ involves ‘differential reinforcement’ . When we add ‘discriminative control’ to the mix we get a powerful technology for building repertoires of behaviour.

When it comes to building a repertoire of behaviour it makes sense to have a plan. And you can only have a plan once you accept that controlling behaviour is possible, which also means that behaviour is predictable.

In the video clips on animal agility, for example, you will see that chains of behaviour are established using what we know about the 3-term contingency. The final part of a chain might look like Fig. 7.3.

Figure 7.3

When the chain is being built, it takes advantage of reinforcing properties acquired by the SD. That is, the SD can be used as a ‘conditioned reinforcer’ like Fig. 7.4.

Figure 7.4

Eventually you end up with a full chain like Fig. 7.5.

Figure 7.5

In applied work, chains of behaviour are established in a similar fashion, though at times you might use a forward chaining procedure instead of the backward chaining procedure just described. That is, you start with the first link and then add more as you go along (Fig. 7.6)

Figure 7.6

Download the tutorial on Discrete Trial Teaching (DTT) that takes you through some basic steps in managing the establishment of SDs for people who need extra care to be taken in their teaching.

Feel free to adjust the material for your own teaching, but it is not to be sold for profit.

Additional Video Resources