Recognize other terms used for respondent conditioning.
Outline the three phases of respondent conditioning. Define all terms.
Describe and exemplify higher order conditioning.
Contrast appetitive and aversive conditioning.
Contrast excitatory and inhibitory conditioning.
Outline and describe the four temporal presentations of US and NS in respondent conditioning.
Describe the phenomena of pseudoconditioning.
Pavlov and His Dogs
You have likely heard about Pavlov and his dogs but what you may not know is that this was a discovery made accidentally. Ivan Petrovich Pavlov (1849-1936; 1927), a Russian physiologist, was interested in studying digestive processes in dogs in response to being fed meat powder. What he discovered was the dogs would salivate even before the meat powder was presented. They would salivate at the sound of a bell, footsteps in the hall, a tuning fork, or the presence of a lab assistant. Pavlov realized there were some stimuli that automatically elicited responses (such as salivating to meat powder) and those that had to be paired with these automatic associations for the animal or person to respond to it (such as salivating to a bell). Armed with this stunning revelation, Pavlov spent the rest of his career investigating the learning phenomenon and won a Nobel Prize in 1904 for his work.
The important thing to understand is that not all behaviors occur due to reinforcement and punishment as operant conditioning says. In the case of respondent conditioning, antecedent stimuli exert complete and automatic control over some behaviors. We saw this in the case of reflexes. When a doctor strikes your knee with that little hammer it extends out automatically. You do not have to do anything but watch. Babies will root for a food source if the mother’s breast is placed near their mouth. If a nipple is placed in their mouth, they will also automatically suck, as per the sucking reflex. Humans have several of these reflexes, though not as many as other animals, due to our more complicated nervous system.
Respondent Conditioning Described
Respondent conditioning occurs when we link or pair a previously neutral stimulus with a stimulus that is unlearned or inborn, called an unconditioned stimulus. Note that this form of learning also goes by the name classical conditioning or Pavlovian conditioning in honor of Ivan Pavlov.
Respondent conditioning is best described as occurring in three phases: pre-conditioning, conditioning, and post-conditioning. See Figure 4.1 for an overview of Pavlov’s classic experiment.
Let’s define terms first. The term conditioning means learning. So pre-conditioning is before learning occurs, conditioning is during learning or the acquisition of the relationship between the two stimuli, and post-conditioning is after learning has occurred. If we say something is un-conditioned it is not learned. Going back to our earlier philosophical discussion, this is learning that is innate or present at birth. Also keep in mind that the stimulus is what is sensed in the world through vision, hearing, smell, taste, or touch. The response is the behavior that is made. Making sure you have the terms straight will help you to understand respondent conditioning easier.
Pre-conditioning. Notice that pre-conditioning has both an A and a B panel. All this stage of learning signifies is that some knowledge is already present. In Panel A, the taste of food makes a dog salivate. This does not need to be trained and is the relationship of an unconditioned stimulus (US) yielding an unconditioned response (UR). The association occurs naturally. In Panel B, we see that a neutral stimulus (NS) yields nothing. Dogs do not enter the world knowing to respond to the ringing of a bell (which it hears).
Conditioning. Conditioning is when learning occurs. Through a pairing of a neutral stimulus and an unconditioned stimulus (bell and food, respectively) the dog will learn that the bell ringing (NS) signals food coming (US) and salivate (UR). The key is that the NS is presented just before the US which yields a UR (in most cases; more on that in a bit).
Post-conditioning. Post-conditioning, or after learning has occurred, establishes a new and not naturally occurring relationship of a conditioned stimulus (CS; previously the NS) and conditioned response (CR; the same response). So, the dog now reliably salivates at the sound of the bell because he expects that food will follow, and it does. If it doesn’t, the response ends or extinguishes as you will see later.
Figure 4.1. Pavlov’s Classic Experiment
Let’s now clearly define our terms:
Unconditioned stimulus — The stimulus that naturally elicits a response.
Unconditioned response —The response that occurs naturally when the US is present.
Neutral stimulus — A stimulus that causes no response.
Conditioned stimulus — The initially neutral stimulus that has been associated with a naturally occurring stimulus to bring about a response.
Conditioned response — The response which is elicited by a CS, though it is not the same as the UR. This response is usually weaker than the UR (the dog salivates to the bell, though it does not do it as much as it does to the sight/smell/taste of the food).
Note to Student: Be sure you not only understand these terms but the acronyms used to represent them. I will use the shorthand the rest of the way through this module and in other places in the book.
To fully understand respondent conditioning, know that the pairings of an NS and US each represent a single trial, called the conditioning trial. The period between conditioning trials is called the intertrial interval. The period between the presentation of the NS and then the US (Panel C) within a conditioning trial is called the interstimulus interval.
The entire process of conditioning, to include when we first make the association between an NS and US to its strengthening over time through repeated pairings, is called acquisition. It is likely not surprising to learn that conditioning occurs quicker if the US is more intense. We will be more motivated to learn to associate making an incorrect response with shock if we receive 150 volts compared to 25 volts.
Conditioning is also more effective when the trials are spaced rather than massed (Menzel et al., 2001). For instance, spacing the trials 5 minutes apart is more effective than spacing them 25 seconds apart. One explanation for this is that we have time to rehearse the CS and US in memory during the intertrial interval and if a new trial occurs too soon, it could interfere with rehearsal (Wagner, Rudy, & Whitlow, 1973).
And we can determine how good the learning is if we toss in a test trial occasionally in which the NS is presented alone to see if it elicits the response (UR/CR; ring the bell alone and see if salivation occurs). We can also wait to see if after the presentation of the NS (bell) and before the US appears (sight of food) if the UR/CR appears on its own (salivation). In other words, does the response occur during the interstimulus interval?
Conditioning and its Different Forms
It is worth noting that the conditioning procedure described in the preceding section on Pavlov is not the only form it can take. In this section, we will discuss a type of layered conditioning, conditioning based on the event being something we desire or want to avoid, conditioning based on the presentation or removal of the US, and finally temporal factors that can produce unique conditioning procedures.
Higher order conditioning.
Sometimes, a stimulus that is associated with a CS (formerly the NS) becomes a CS itself and elicits the CR. We call this higher order conditioning, and each level of conditioning is referred to as first, second, third, etc. order conditioning. So how might this work?
Being assaulted (US) will elicit fear (UR). A person wearing a ski mask would alone not cause any response (it is an NS1). If though, you pair the person wearing the ski mask (NS1) with the assault (US) which causes fear (UR), then the sight of a person wearing a ski mask (CS1) will elicit fear (CR). Keep in mind that with the stimuli, you see a person wearing a ski mask and feel the effects of the assault (touch or pain receptors in the skin will be activated). This is first-order conditioning (not to be confused with the training of First Order stormtroopers in Star Wars) and in this example involves a person being associated with fear.
But what if the assault occurred in an alley in your neighborhood? Now the alley (NS2) is paired with the person wearing the ski mask (CS1) which causes fear (CR), and post-conditioning shows that the alley (CS2) causes fear (CR). This is second-order conditioning and involves a location being associated with fear.
Could the time of day be a factor too? What if the mugging occurred at night? If night (NS3) is paired with the alley (CS2) which causes fear (CR), then being outside at night (CS3) could lead to fear (or at least some anxiety; CR). This would be third-order conditioning and now involves a time of day being associated with fear.
Fear was originally elicited by being assaulted. Through higher order conditioning, it was also elicited by the sight of a ski mask, being in an alley, and being outside at night. The fear reaction becomes weaker across the conditioning of these additional NS, such that our response to being outside at night could be better classified as anxiety and not so much the bona fide fear felt while being assaulted (and likely for a time afterward) which suggests that the response is strongest to the US and becomes weaker across CS1, CS2, and CS3.
Appetitive and aversive conditioning.
Recall from Section 2.1.3 that appetitive stimuli are those that an organism desires and seeks out while aversive stimuli are readily avoided. In respondent conditioning, the US could be an appetitive or aversive stimulus. For instance, in appetitive conditioning, the US would be something desirable such as candy which makes us happy. Other examples could include water, food, sex, or drugs. In aversive conditioning, the stimulus is not pleasant and could include extreme temperatures, a painful sting such as from a wasp or a bite from a dog, electric shock, or something that does not smell nice. It would not be surprising to learn that conditioning occurs relatively fast when aversive US are involved. Since these stimuli could harm or kill us, learning to avoid them is adaptive and aids our survival.
Excitatory and inhibitory conditioning.
Our discussion so far has included examples in which the NS is associated with the presentation of the US, called excitatory conditioning. For Palov’s dogs, they associated the ringing of a bell (NS) with the presentation of the food (US) which caused their salivation (UR). Eventually, salivation (CR) occurred to just the ringing of the bell (CS).
Interestingly enough, the absence of the US could be associated with an NS too, in a process called inhibitory conditioning. Go back to our example for higher conditioning. A person wearing a ski mask is an excitatory CS for fear but seeing someone wearing such a mask during the daytime leads to an inhibition of fear. It being day indicates a safe interval and we will not be overly concerned about ski masks. We have only ever been assaulted at night. The excitatory CS is expressed as CS+ and the inhibitory CS as CS-.
Temporal factors affecting conditioning.
In the previous section we saw that generally, the US is presented after the NS though the NS could be followed by the absence of an US. These examples have also always presented the NS before the US, but this is not necessary in all cases.
First, delay conditioning involves the presentation of the NS before the US, but the NS overlaps with the US for a short period of time. In the case of Pavlov’s experiment, the bell would ring for say 10 seconds, then the food would enter the room, and then the bell would end 5 seconds after this. The ISI (interstimulus interval) should be relatively brief to use this procedure.
What if we present the NS well ahead of the US in time? Let’s say we ring the bell for 10 seconds and then there is a 5-second gap before the food enters the room. The NS and US do not overlap. This is the basis of trace conditioning and the trace is a memory that we have to access. The organism will need to remember that the NS occurred before the US to make the association, or that the bell rang before the food came in. The period of time between the NS terminating and the US beginning is called the trace interval and ideally should be short, or a few seconds.
The NS and US could occur at the same time such as in simultaneous conditioning. As you might expect, conditioning in this procedure is poor since the NS does not predict the occurrence of the US. They occur simultaneously. The bell would ring as the food enters the room. The bell-ringing does not lead to an expectation that food will come shortly, which aids with learning the association.
Finally, the US could come before the NS in a procedure called backward conditioning. The US would occur first and last for a few seconds with the NS starting near the end of this time. Hence, the NS and US co-occur for a short period of time. Of the four methods, backward conditioning is the least effective for excitatory conditioning though it could lead to inhibitory conditioning. Consider a shock paradigm in which a rat is given a shock (US) and then near the end of the shock a light is turned on (NS). The light (NS) would signal the end of the shock (US) and serve as a safety signal. Hence, the NS would become a CS-.
How Do You Know if Learning Occurred?
A cardinal feature of science is to verify that any change in your variable of interest (the DV) is caused by the treatment or manipulation (the IV). It could be that the elicited response was not actually caused by the NS/CS and so a product of learning or conditioning, but was caused by sensitization instead, called pseudoconditioning.
Let’s say you were working with turtles and presented them with a tone (the NS) followed by tapping on the shell (US) which resulted in the turtles withdrawing into their shells (UR). With a few such pairings, the tone (CS) would lead to withdrawing into shells (CR). So the tone has been associated with tapping, right? Possibly, but let’s say in addition to the tone we also flash a light. The turtles also withdraw into their shells at the presentation of this stimulus. In the case of sensitization, repeated presentation of a stimulus leads to an increase in how strong the response is. It can also lead to other stimuli eliciting the same response as in the case of the bright light and tone both eliciting the withdraw into shell response.
To know if the effect on the behavior you are seeing is due to conditioning and not sensitization, a simple adjustment can be made — the inclusion of a control group. The experimental group would have the tone and tap paired together resulting in a withdrawal response. The control group would have the tone played and then the tap made far apart in time. Now when the tone is presented to each group alone, the experimental group would have a strong withdrawal into shell response while the control group may have the same response, but it would be weak. The intensity of the response, or in this case it being stronger in the experimental rather than control condition, indicates conditioning has truly occurred. There is no pseudoconditioning in other words.
