Before introducing you to the various types of Pavlovian Conditioning, a few additional points need to be made. First, Pavlovian Conditioning is not simply stimulus substitution (Catania, 1998, p. 201). In both of the previous examples, the form or topography of the UR and the CR appears to be very similar, but this is not always the case. For example, the effect of heroin, a US, is to produce an elevated pain threshold (UR); an event that typically precedes heroin use can become a CS, but its effect is to lower pain threshold (CR).
Second, consider the following experiment. A mild shock to the foot (US) elicits foot withdrawal (UR). A buzzer (NS) reliably precedes the mild shock (US). Later, it is discovered that the buzzer alone produces foot withdrawal. Is this Pavlovian Conditioning? It may not be. In Pavlovian Conditioning, the CS elicits the CR because the CS has been paired with the US. In our example, we must be able to make the case that foot withdrawal occurs to the buzzer due to the earlier pairing of the buzzer with the mild shock. In an experiment similar to this one (Wickens & Wickens, 1940), the buzzer elicited foot withdrawal even when it had not been previously paired with the mild shock; rather, earlier "training" had simply involved presenting the mild shock by itself. This is an example of pseudoconditioning: a NS may produce a response because of repeated presentations of a US alone or repeated presentations of the NS alone, independent of any relation between the two stimuli (Underwood, 1966, pp. 386-387; see also Catania, 1998, p. 204).
Third, Pavlovian Conditioning is to be distinguished from Operant Conditioning. Pavlovian Conditioning is based on elicited responding. Behavior change occurs due to the presentation of an antecedent stimulus; behavior change is not dependent on the past consequences of the behavior. For example, Pavlov's dogs salivated to the food (antecedent stimulus) regardless of any consequence salivating might have produced. Even had Pavlov arranged a consequence for this elicited behavior, such as omitting scheduled food presentations, salivating to food would have continued (Catania, 1998, p. 211).
Operant Conditioning is based on emitted responding. Behavior change occurs because of the consequences produced by that behavior. Consider an emitted behavior such as lever pressing. Suppose a rat's lever presses remove the opportunity to eat food. Lever pressing will likely decrease in frequency, not because of any triggering antecedent stimulus, but rather because a stimulus change (food removal) had reliably followed its occurrence. (Review the Positive Reinforcement Tutorial to learn more about a type of Operant Conditioning.)
The final point concerns terminology. Consider the following scenario that extends the Little Albert experiment. Suppose that Albert visits his friend, who invites him down into the basement to see something really cool. His friend, unaware of Albert's phobia, shows Albert his pet white rat, causing Albert to cry. Consequently, Albert is fearful the next time he is at his friend's house and sees the basement door. In this case, because the basement door preceded Albert seeing the white rat, the basement door alone came to elicit fear. How are we to label the terms in this example? According to our definition, the white rat is not an US - Albert's fear of it does depend on prior learning. The white rat had been a neutral stimulus prior to its pairing with the loud noise. What we have here is an case of higher-order conditioning, a topic beyond the scope of this exercise (see Mazur, 1994, pp. 76-77). For our purposes, we will refer to any stimulus that reliably elicits a response prior to any subsequent pairing of it with another stimulus as the US. Thus, in this example, the white rat (US) elicits crying (UR); the basement door (NS) signals the white rat (US); as a result, the basement door (CS) elicits crying (CR).