Ch. 5 Flashcards
Stimulus-Substitution Theory
(Pavlov’s)
the CS acts as a substitute for the US.
Pavlov was a physiologist who believed that classical conditioning was an indirect way of studying neurological processes in the brain.
Thus, he often made inferences about the kinds of neurological processes that are activated during conditioning.
Pavlov’s notions about the kinds of neurological processes underlying classical conditioning are now considered to be incorrect.
The actual processes involved are known to be considerably more complex than he presumed.
Nevertheless, this does not negate all aspects of Pavlov’s theory.
For example, consider the notion that the conditioned stimulus (CS) is somehow a direct substitute for the US.
In at least some cases, it seems as though animals do react to the CS as if it were the US.
The dog salivates to the light just as it does to food.
More importantly, the dog may even approach the light and start to lick it, as though pairing the light with the food resulted in the light being perceived as edible (Pavlov, 1941).
This sort of phenomenon, now known as sign tracking,
Stimulus-Substitution Theory examples
Pavlov claimed that presentation of a US, such as food, activates an area of the cerebral cortex (the outermost layer of the brain) that is responsible for sensing the occurrence of that event.
Activation of this “food center” in the brain in turn activates another part of the cortex (the “salivation center”) that produces the unconditioned response of salivation.
Food - >Activates food center in cortex - > Activates salivation center in cortex - >
Salivation
Pavlov also believed that the presentation of a neutral stimulus, such as a light, activates another area of the cortex responsible for detecting that type of stimulus. when the light is presented just before the food, a connection is formed between the area of the cortex activated by the light and the area activated by the food.
As a result, activation of the light center of the cortex also activates the food center of the cortex, which in turn produces salivation.
In other words, Pavlov believed that the presentation of the light set in motion the following sequence of events:
Light -> Activates light center in cortex -> Activates food center in cortex -> Activates salivation center in cortex -> Salivation
A particular difficulty for stimulus-substitution theory
is that if the CS is acting as a substitute for the US, then the CR should always be the same, or at least highly similar, to the UR.
Although this is often the case, it sometimes is not the case; in fact, there are sometimes substantial differences between the CR and the UR.
For example, a rat that receives a foot shock (the US) will probably jump (the UR).
However, if it sees a light (CS) that has been paired with a foot shock, it will often freeze (the CR).
Why would the rat jump in one instance and freeze in the other?
An examination of the rat’s natural response to danger gives us a clue.
If a rat is attacked by a snake, jumping straight up (and rats can really jump!) may cause the snake to miss.
On the other hand, if a rat detects a snake in the vicinity, tensing its muscles and freezing will minimize the possibility of being detected or, if the rat is attacked, will enable it to jump quickly.
This suggests that the purpose of the CR, rather than merely being a version of the UR, is to ready the organism for the occurrence of the US which brings us to our next theory of conditioning.
preparatory-response theory of conditioning
the purpose of the CR is to prepare the organism for the presentation of the US.
The dog salivates to the tone to get ready for food, and the rat freezes in response to the light to get ready for the
shock.
Note that in one case, the preparatory response is quite similar to the UR, whereas in the other case it is quite different.
Thus, a major advantage of preparatory-response theory is that, unlike stimulus-substitution theory, it accounts for situations in which the CR and the UR are sometimes different.
preparatory-response theory of conditioning example
example of conditioned preparatory-responses involves cases in which the CR seems to be the exact opposite of the original UR. This often occurs with drug reactions.
Heroin is an opiate drug,It is an analgesic and a narcotic, meaning that it kills pain and makes you sleepy. It is highly addictive, and users report that the high associated with heroin is very powerful.
Imagine that someone always injects heroin in the presence of certain environmental cues,
such as in a particular room. Thus, regularly injecting heroin in a particular setting involves the following sequence of events:
Heroin cues: (NS) Heroin (US) - BP decreases (UR)
If this were a normal conditioning procedure, one might expect that the heroin-related cues would eventually become a CS that will itself elicit a decrease in blood pressure. But in reality, the opposite occurs.
With repeated drug use, the presence of the heroin-related cues elicits not
a decrease in blood pressure, but an increase in blood pressure!
Heroin cues: (CS) - BP Increases (CR)
Similar to opponent-process theory of emotion, certain stimuli can elicit both a primary response (the a-process) and a
compensatory response (the b-process). Thus gives way to compensatory-response model.
compensatory-response model
a CS that has been repeatedly associated with the primary response (the a-process) to a US will eventually come to elicit a compensatory response (the b-process).
Why would this type of compensatory conditioning occur?
Remember how in the opponent-process theory, the compensatory reactions to a US serve to maintain a state of homeostasis (internal balance).
If these compensatory reactions start occurring before the US is presented, they will be even more effective in minimizing the disturbance produced by the US.
Conditioned compensatory responses therefore constitute an extreme form of preparatory response to certain environmental events.
compensatory-response model example
Repeatedly injecting heroin does not simply elicit a response, but instead sets in motion a chain of events.
The heroin directly elicits an immediate decrease in blood pressure (the a-process) that in turn
elicits a compensatory increase in blood pressure (the b-process).
Heroin - BP decreases (a-process) - BP Increases (b-process)
In terms of stimuli and responses, the heroin is a US that naturally elicits a decrease in blood pressure, and the decrease in blood pressure is itself a US that naturally elicits an increase in blood pressure.
Therefore, the decrease in blood pressure is both an unconditioned response (UR) to heroin and an unconditioned stimulus (US) that elicits a compensatory increase in blood pressure.
Heroin (US) - BP decreases (UR/US) - BP Increases (UR)
Notice that there are two US in this sequence, with which the cues in the environment could potentially become associated: one US is the heroin and the other is the decrease in blood pressure that results from the heroin.
What happens in compensatory conditioning is that the heroin-related cues, such as being in a certain room, become associated not with the heroin itself but with the primary response to heroin that is, with the decrease in blood pressure.
As a result, these cues eventually come to elicit the compensatory reaction to that response. So the actual conditioning that takes place with heroin is as follows:
Heroin cues (NS): BP decreases (US) - BP Increases (UR)
Heroin cues (CS) - BP Increases (CR)
if the compensatory reaction to the heroin (an increase in blood pressure) can be elicited just before the injection of heroin, then the immediate physical reaction to the heroin (the decrease in blood pressure) will be effectively moderated.
In this sense, a conditioned compensatory response allows the body to prepare itself ahead of time for the onslaught of the
drug.
Unfortunately, for many people trying to kick a habit, whether it is alcohol, cigarettes, or heroin, it is often not possible to completely avoid all cues associated with the drug.
For this reason, modern treatments for drug addiction often include procedures designed to extinguish the power of such cues.
For example, someone attempting to quit smoking may be required to remain in the presence of cigarettes for a long period of time without smoking.
Repeated presentations of the CS (the sight of the cigarettes) in the absence of the US (nicotine
ingestion) should result in weaker and weaker CRs (cravings for a smoke).
Of course, this process can initially be very difficult, It therefore requires careful management; however, once accomplished, it can significantly reduce the possibility of a relapse.
The compensatory-response model also has implications for drug tolerance
For example, if you have a habit of always drinking in a particular setting, then the various cues in that setting people greeting you as you walk in the front door of the bar; the stool you always sit on- become CSs for the effect of alcohol.
The presence of these CSs will initiate physiological reactions that compensate for the alcohol you are about to consume.
As a result, in the presence of these CSs, you should have greater tolerance for alcohol than you would in their absence.
Amount of alcohol you consume is not, by itself, a reliable gauge for determining how intoxicated you are.
Exceptions to the typical compensatory reactions to a CS.
Stimuli associated with drug use sometimes elicit drug like reactions rather than drug- compensatory reactions. In other words, the stimuli become associated with the primary response to the drug rather than the compensatory response.
There is also evidence that stimuli associated with drug use sometimes elicit both compensatory responses in one system of the body and drug-like responses in another.
Thus, the circumstances in which conditioning results in drug-like reactions versus drug-compensatory reactions are complex and not entirely understood
Rescorla-Wagner Theory (1972).
Proposes that a given US can support only so much conditioning, and this amount of conditioning must be distributed among the various CSs that are present.
Another way of saying this is that there is only so much associative value available to be distributed among the cues associated with the US.
One assumption is that stronger stimuli (USs and CSs) support more conditioning than do weaker stimuli.
The theory can also been interpreted in more cognitive terms.
To say that a CS has high associative value is similar to saying that it is a strong predictor of the US, or that the subject strongly “expects” the US whenever it encounters the CS.
Rescorla-Wagner Theory example
For example, the use of a highly preferred food as the US produces a stronger conditioned response of salivation than does a less preferred food.
Imagine, for example, that a tone paired with a highly preferred food (say, steak) elicits a maximum of 10 drops of saliva, while a tone paired with a much less preferred food (say, dog food) elicits only 5 drops of saliva.
If we regard each drop of saliva as a unit of associative value, then we could say that the highly preferred food supports a maximum associative value of 10 units, while the less preferred food supports a maximum associative value of 5 units.
We can use the following format to diagram the changes in associative value (we will assume the highly preferred food is the US):
Tone (V = 0 ): Food (Max = 10 ) - Salivation
Tone ( V = 10 ) - Salivation
The letter V will stand for the associative value of the CS (which at the start of conditioning is 0). The term Max will stand for the maximum associative value that can be supported by the US once conditioning is complete.
V as the number of drops of saliva the tone elicits-0 drops of saliva to begin with and 10 drops once the tone is fully associated with the food-and Max as the maximum number of drops of saliva that the tone can potentially elicit if it is fully associated with the food.
Now suppose that a compound stimulus consisting of a tone and a light are repeatedly paired with the food, to the point that the compound stimulus obtains the maximum associative value.
[Tone + Light] ( V = 0 ): Food (Max = 10 ) - Salivation
[Tone + Light] ( V = 10 ) - Salivation
This associative value, however, must somehow be distributed between the two component members of the compound.
If tone is slightly more salient than the light, then the tone might have picked up six units of associative value while the light picked up only four units. In other words, when tested separately, the tone elicits six drops of saliva while the light elicits four.
Tone ( V = 6 ) - Salivation
Light (V = 4 ) - Salivation
If the tone were even more salient than the light-for example, it was a very loud tone and a very faint light-then overshadowing might occur, with the tone picking up nine units of associative value and the light only one unit:
[Loud tone + Faint light] ( V = 0 ): Food ( Max = 10 ) - Salivation
Loud tone ( V = 9 ) - Salivation
Faint light ( V = 1 ) - Salivation
The loud tone now elicits nine drops of saliva (a strong CR) while the faint light elicits only one drop of saliva (a weak CR).
Rescorla-Wagner explanation for the overshadowing effect
is that there is only so much associative value available (if you will, only so much spit available) for conditioning, and if the stronger stimulus in the compound picks up most or all of the associative value, then there is little or no associative value left over for the weaker stimulus.
Rescorla-Wagner theory readily explains conditioning situations involving compound stimuli. Take, for example, a blocking procedure.
One stimulus is first conditioned to its maximum associative value:
Tone (V = 0 ): Food ( Max = 10 ) - Salivation
Tone (V = 10 ) - Salivation
This stimulus is then combined with another stimulus for further conditioning trials:
[Tone + Light] ( V = 10 + 0 = 10 ): Food (Max = 10 ) - Salivation
But note that the food supports a maximum associative value of only 10 units, and the tone has already acquired that much value.
The light can therefore acquire no associative value because all of the available associative value has already been assigned to the tone.
Thus, when the two stimuli are later tested for conditioning, the following occurs:
Tone ( V = 10 ) - Salivation
Light ( V = 0 ) - No salivation
Rescorla-Wagner theory also leads to some counterintuitive predictions.
Consider what happens if you first condition two CSs to their maximum associative value and then combine
them into a compound stimulus for further conditioning.
For example, suppose we condition a tone to its maximum associative value, as follows:
Tone (V = 0 ): Food ( Max = 10 ) - Salivation
Tone ( V = 10 ) - Salivation
and then do the same for the light:
Light (V = 0 ): Food (Max = 10 ) - Salivation
Light ( V = 10 ) - Salivation
We now combine the tone and the light into a compound stimulus and conduct further conditioning trials:
[Tone + Light] ( V = 10 + 10 = 20 ): Food (Max = 10 ) - Salivation
Note that the tone and the light together have 20 units of associative value (10 for the tone and 10 for the light). However, the maximum associative value that can be supported by the food at any one moment is only 10 units. This means that the associative value of the compound stimulus must decrease to match the maximum value that can be supported by the US.
Thus, according to the Rescorla-Wagner theory, after several pairings of the compound stimulus with food, the total associative value of the compound stimulus will be reduced to 10:
[Tone + Light] ( V = 10 ) - Salivation
This in turn means that when each member in the compound is tested separately, its value also will have decreased. For example:
Tone (V = 5 ) - Salivation
Light ( V = 5 ) - Salivation
Thus, even though the tone and light were subjected to further pairings with the food, the associative value of each decreased (i.e., each stimulus elicited less salivation than it originally did when it had been conditioned individually).
This is a counterintuitive result in that one would normally expect further pairings between a CS and US to either maintain or strengthen
overexpectation effect
which is the decrease in a CR that occurs when two separately conditioned CSs are combined into a
compound stimulus for further pairings with the US.
It is as though presenting the two CSs together leads to an “overexpectation” about what will follow. When this expectation is not fulfilled, the subject’s expectations are modified downward. As a result, each CS in the compound loses some of its associative value.
Watson and Rayner’s “Little Albert”
The experimenters thus concluded that the loud noise was an unconditioned stimulus that elicited a fear response (or, more specifically, a startle reaction), whereas the other objects, such as the rat, were neutral stimuli with respect to fear:
Watson and Rayner (1920) paired the white rat (NS) with the loud noise (US).
The rat was presented to Albert, and just as his hand touched it, the steel bar was struck with the hammer.
In this first conditioning trial, Albert “jumped violently and fell forward, burying his face in the mattress.
He did not cry, however” He reacted similarly when the trial was repeated, except that this time he began to whimper.
The conditioning session was ended at that point.
The next session was held a week later. At the start of the session, the rat was handed to Albert to test his reaction to it.
He tentativelv reached for the rat, but quickly withdrew his hand after touching it.
Since, by comparison, he showed no fear of some toy blocks that were handed to him, it seemed that a slight amount of fear conditioning to the rat had occurred during the previous week’s session.
Albert was then subjected to further pairings of the rat with the noise, during which he became more and more fearful.
Finally, at one point, when the rat was presented without the noise, Albert “began to crawl so rapidly that he was caught with difficulty before reaching the edge of the table”
Albert’s reaction was interpreted by Watson and Rayner as indicating that the rat had indeed come to elicit a conditioned fear response as a result of its association with the noise.
In subsequent sessions, during which Albert occasionally received additional conditioning trials, he reportedly showed not only a fear of the rat but also of objects that were in some way similar to the rat.
Albert’s fear seemed to have generalized to objects that were similar to the original CS.
His apparent fear of the rat, and his generalized fear of similar objects, persisted even following a 30-day break, although the intensity of his reactions was somewhat diminished.
At that point, Albert left the hospital, so no further tests could be conducted.
Watson and Rayner were also unable to carry out their original plan of using behavioral procedures to eliminate Albert’s newly acquired fear response.
Thus, although Watson and Rayner (1920) speculated about the possibility of Albert growing up to be a neurotic individual with a strange fear of furry objects, there are strong reasons to believe that this did not occur.
And even if he had acquired a conditioned fear response, additional factors are usually needed for such fears to develop into a true phobia.
Watson and Rayner’s “Little Albert” diagram
Loud noise (US) -> Fear (indicated by startle reaction) (UR)
Rat (NS) - >No fear (-)
Rat (NS): Loud noise (US) - Fear (UR)
Rat (CS) - Fear (CR)