4.2: Properties Governing Respondent Conditioning

Section Learning Objectives

• Define extinction.
• Describe spontaneous recovery in relation to extinction.
• Differentiate stimulus generalization and discrimination.
• Describe sensory preconditioning.
• Describe latent inhibition.
• Define overshadowing.
• Define blocking.
• Explain the use of occasion setters.

Extinction and Spontaneous Recovery

Once an association between the NS and US has been established resulting in the NS becoming a CS, is there a way to break this association? The answer is yes, and respondent extinction involves the CS no longer being paired with the US leading to no response when the CS is presented again. For instance, the sound of a bell ringing (CS) is not followed by food (US) as the animal has come to expect and predict, and so eventually the dog stops salivating (the CR) when the bell sounds.

This property leads us to wonder if the broken association of the CS and US is permanent. The answer is no and eventually, the bell will ring making the dog salivate. If no food comes, the behavior will not continue. The organism may make the response a few more times with the strength of the response weakening each time until eventually it ends. If food comes, the salivation response will be re-established. This property is called spontaneous recovery and is when the CS elicits the CR after extinction has occurred. The association between CS and US is re-established relatively quickly once the pairing is made again.

Stimulus Generalization and Discrimination

When a number of similar CS or a broad range of CS elicit the same CR, stimulus generalization is said to have occurred. An example is the sound of a whistle eliciting salivation the same as the sound of a bell, both detected via audition. As you would expect, the more similar the new stimulus is to the original CS, the stronger the response will be. If a child was conditioned to be afraid of white rats, we would expect the response to be stronger if made in the presence of a white mouse rather than a German shepherd.

When the CR is elicited by a single CS or a narrow range of CSs, stimulus discrimination is said to have occurred. Teaching the dog to not respond to the whistle but only to the bell, and just that type of bell, is an example. Other bells would not be followed by food, eventually leading to the extinction of the erroneous association. Teaching an organism to make such discriminations is called discrimination training.

Sensory Preconditioning

There are situations in which a stimulus becomes a CS, making other stimuli it was paired with likely candidates to become a CS in the future too. This is called sensory preconditioning. Years ago, I worked for the National Institutes of Health where I did learning and memory experiments on rats and mice. Though when I started the rats in particular did not scare me, one fateful March day I injected a rat in the wrong spot, hurt it (unintentionally, of course), and it promptly turned around and took a piece of my hand with it. The bite of course elicits fear in keeping with a US-UR relationship, and for me the rat was an NS. Through that experience, and some close calls after that, I came to associate rats (CS) with fear (CR) when injecting them. Before all this, I associated rats (NS1) with their home cage where they lived (NS2) from which I took them out to handle. After the bite, I became afraid (CR) of handling rats when injecting them again (CS1), and when handling them in the home cages (CS2). Animals can detect fear, so this was not a good development, but also my confidence declined.

What was the end result? After a few months, I no longer worked with rats. This was not because of the bite incident. In keeping with what we learned about extinction above, eventually having numerous attempts injecting rats and not being bitten, the fear would have extinguished. That is what happened. Unfortunately, when the rat bit me he intensified my allergies that I did not realize were to the rats and mice I worked with. I believed they were just normal seasonal allergies and worse that year than past years, which did happen from time-to-time. I went about my business, ignorant of just how close to going into anaphylactic shock I truly was. So exit mice and rats. Enter fruit flies as my research subject of choice for learning and memory experiments from that point forward. I did that for about a year and finally had enough.

Latent Inhibition

The property of latent inhibition states that it is easier to condition a novel stimulus than a familiar one (Lublow, 1973). If you are using music as an NS, use a song you are unfamiliar with (or your subjects are) such as a Barry Manilow song, rather than one they know and listen to often such as Lady Gaga or Taylor Swift. So in your experiment, the song (NS) is followed by food (US) which elicits salivation (UR) in the person (much like one of Pavlov’s dogs). If you use that Taylor Swift song as the NS, salivation is not likely to occur, assuming the participants have heard it numerous times already. If you use the Barry Manilow song (CS), they will likely make the salivary response as expected (CR). The benefit of latent inhibition is that we do not form associations between CRs and repetitive stimuli in our environment that are linked by mistake or coincidentally.

Complementary to latent inhibition is the concept of US preexposure effect or exposure to a US before conditioning occurs which can make subsequent conditioning more difficult (Randich & LoLordo, 1979). Hence, the more preexposure an organism has to a US, the worse learning is later and habituation may be the culprit here. Being exposed repeatedly to the US before conditioning, the organism may habituate to it at least to some degree, making conditioning more difficult.

Overshadowing and Blocking

There are times when we are presented with two or more stimuli simultaneously, called a compound stimulus. We might, for instance, be presented with a light and a sound at the same time. In what is called overshadowing (Pavlov, 1927), two neutral stimuli are presented at the same time and the more salient of the two becomes a CS. Let’s say a green light and high-pitched tone were both presented at the same time and paired with the US of food which elicits salivation (UR). Which part of the compound would become the CS and elicit salivation? It appears the tone which was high-pitched would stand out more than a mere green light. But what if this green light was neon green or flashing and the tone was relatively faint and monotone? Now the light would be the more relevant or salient stimulus and become conditioned as the CS. The more salient part of the compound stimulus causes the less salient part to elicit little to no response, or remain as an NS.

In blocking (Kamin, 1969; 1968), the compound stimulus is composed of an NS and a CS and the established CS interferes with learning a new CS relationship. This differs from overshadowing which has two NS as part of the compound stimulus differing only in terms of salience. Let’s use the example above. Say you presented a dog with a tone (NS) and food (US) leading to salivation (UR). The dog will learn that when it hears the tone (CS) food is coming and will salivate (CR). What if now you present the tone (CS) with a green light several times (NS; note that the tone and light are a compound stimulus and presented simultaneously) and pair them with food (US), which causes salivation (UR). By doing this, you are trying to teach the dog to salivate to the green light, but when the stimuli are presented separately, the tone (CS) elicits salivation (CR) while the green light (NS) causes no response. The former learning has blocked the new learning.

Occasion Setting

The context in which learning occurs is also important. Occasion setters are stimuli that help an organism determine if the CS will be followed by a US leading to the CR. In other words, the CS elicits a CR only when the feature is present, called a feature-positive occasion setter. This feature may also indicate when a CS will not be followed by the US, called a feature-negative occasion setter (Palmatier, 2014). Consider the example of a bell (NS) being followed by food (US), which elicits salivation (UR). What if the food only comes out if the bell rings when a white light over it turns on (OS or occasion setter). If the light is on (OS) when the bell rings (NS), food comes out (US), leading to salivation (UR). If the light is off when the bell rings (NS), no food comes out, which will not elicit the behavior of salivation. We can test whether the discrimination was made by turning the light on (OS) when the bell rings (CS), which leads to salivation (CR; and the correct response) and then not having the light on when the bell rings, which should cause no response in the organism. If this is true, conditioning and the discrimination was learned. The animal now knows that food will only come out when the light is turned on during the ringing of the bell and salivate.