Chapter 3 Classical Conditioning: Foundations
Chapter 3 Classical Conditioning: Foundations The Early Years of Classical Conditioning -
Studies of classical conditioning began with the work of Russian physiologist Pavlov. o Started when he noticed that dogs created psychic secretions every time they were presented food. Discoveries of Vul’fson and Snarskii -
-
-
Performed the first systematic studies of classical conditioning. Vul’fson focused on salivary glands. o Studied salivary responses to various substances placed in the mouth: dry food, wet food, sour water and sand for example. o If they had the food placed in their mouth repeatedly, the mere sight of these foods was enough to make them salivate. Snarskii extended the studies with the use of artificial substances. o Used different colors to color bowls of water. Some substances produced distinctive taste and texture sensations in the mouth, known as orosensory stimuli. First time only the feeling of sand in dogs mouth elicited salivation, however after the sand had been placed in the mouth several times the sight of sand also came to elicit salivation. o The association of one feature of an object with another is called object learning. To study the mechanisms of associative learning, the stimuli to be associated have to be manipulated independently of one another. o Pavlov used procedures in which the stimuli to be associated came from different sources.
The Classical Conditioning Paradigm -
Involves two stimuli: one of these stimuli is a tone or a light. The other stimulus is food or the taste of sour solution placed in the mouth. In contrast to the light or tone, the food or sour taste elicits vigorous salivation even the first time it is presented. o The light or tone referred to as conditional stimulus b/c the effectiveness of this stimulus in eliciting salivation depended pairing it several times with the presentation of food. o The food or sour taste was called the unconditioned stimulus b/c its effectiveness in eliciting salivation did not depend on any prior training. o The salivation that eventually came to be elicited by the tone or light was called the conditioned response, and the salivation that was
always elicited by the food or sour taste was called the unconditioned response. Experimental Situations -
Pavlov used dogs using salivary-fistula technique. Most contemporary experiments on Pavlovian conditioning are carried out with rats, rabbits, and pigeons.
Fear Conditioning -
Little Albert Experiment o Conditioned rat (CS) with loud abstract noise (US) and made little Albert afraid (CR) of the rat. When rats are scared they become immobile. Two different indirect measures of immobility known as conditioned suppression procedures are: o Lick Suppression Procedure: rat licking a spout of water, then presented with a fear CS (i.e. tone). Now measure how long it takes to lick a certain amount of times. o Conditioned Emotional Response: rats are first trained to press a response lever for food reward in a small chamber. Once the rats are pressing the lever at a steady rate fear conditioning is introduced consisting of a tone or light paired with a brief shock. As they acquire the conditioned fear, they come to suppress their level pressing during the CS. They calculate a ratio and compare to the normal baseline.
Eye Blink Conditioning -
Puff of air to the eye (US) was given when a tone was heard (CS). Eventually began to blink when they heard the tone.
Sign Tracking -
One experimental paradigm that has contributed significantly to modern conceptions of Pavlovian conditioning is the sign tracking (autoshaping). o Not a highly reflexive response like salvation and blinking. Sign tracking is investigated by presenting a discrete, localized visual stimulus just before each delivery of a small amount of food. o I.e. Pigeon in a chamber with a small circular key that could be illuminated and that the pigeons could peck. Periodically the pigeons were given access to the food for a few seconds. The key light was illuminated for 8 seconds immediately before each food delivery. o Instead of using the key light to tell them when they should go the food dish, they started pecking the key itself.
-
o Interesting behavior discovery b/c the key was not needed to gain access to the food. For sign tracking to occur the CS has to be of the proper modality and configuration.
Learning What Tastes Good or Bad -
-
-
-
A conditioned taste aversion is learned if ingestion of a novel flavor is followed by an aversive consequence such as indigestion or food poisoning. In contrast a taste preference may be learned if a flavor is paired with nutritional repletion or other positive consequences. o This can happen in just one trial and the learning can even occur if the illness is delayed several hours after ingestion of the food. In 20% of the cases, individuals believed that their illness was not the result of the food they ate. This indicates that food aversion learning can be independent of rational thought processes and can go against a person’s conclusions about the causes of their illness. I.e. eating before chemotherapy, most people think their loss of appetite is caused by the chemo, but it could be that eating before produced a taste aversion. Taste aversion learning is a result of the pairing of a CS (taste) and a US (drug injection or radiation exposure) in much the same manner as in other examples of classical conditioning. However it does have some special features: o 1. Strong taste aversions can be learned with just one pairing of the flavor and the illness. o 2. Taste aversion learning occurs even of the illness does not occur until several hours after exposure to the novel taste. Evaluative Conditioning: when our evaluation or liking of a stimulus changes by virtue of having that stimulus associated with something we already like or dislike. o Used extensively in the advertisement industry.
Excitatory Pavlovian Conditioning Procedures -
So far we have been learning excitatory Pavlovian conditioning. In excitatory conditioning, organisms learn an association b/w the conditioned and unconditioned stimuli. o As a result of the association, presentation of the CS activates behavioral and neural activity related to the US in the absence of the actual presentation of the US.
Common Pavlovian Conditioning Procedures -
Major factor that determines the course of classical conditioning is the relative timing of the CS and US.
-
Five Common Classical Conditioning Procedures: o Short-delayed Conditioning: Most frequently used. Involves delaying the start of the US slightly after the start of the CS on each trial. Critical feature is that the CS starts each trial and the US is presented after a brief delay. The CS may continue during the US or end when the US begins. o Trace Conditioning: Similar to short delayed procedure. However in trace conditioning, the US is not presented until some time after the CS has ended. This leaves a gap b/w CS and US called the trace interval. o Long-delayed Conditioning: In this case the US is delayed much longer (5-10 mins) than in the short delayed conditioning. Does not include a trace interval. The CS lasts until the US begins. o Simultaneous Conditioning: Presents the CS and US at the same time. o Backward Conditioning: US occurs shortly before, rather than after, the CS.
Measuring Conditioned Responses -
-
To make comparisons among the various procedures, one has to use a method for measuring conditioning that is equally applicable to all the procedures. o This is typically done with the use of a test trial, which consists of presenting the CS by itself. Responses elicited by the CS can then be observed without contamination from responses elicited by the US. Behavior during the CS can be quantified in several ways. o One aspect of conditioned behavior is how much of it occurs. This is called the magnitude of the CR. I.e. Pavlov measured the amount of saliva that was elicited by the CS. o The energy of responding can also be measured by how often the CS elicits a CR. I.e. measure the percentage of trials on which a CR is elicited by the CS. This reflects the likelihood, or probability of responding. o A third aspect of conditioned responding is how soon the CR occurs after the presentation of the CS. This is known, as the latency of the CR. Latency is the amount of time that elapses b/w the start of the CS and the occurrence of the CR.
Control Procedures for Classical Conditioning -
To be certain that a conditioning procedure is responsible for certain changes in behavior; those changes must be compared to the effects of a control procedure. What should the control procedure be? o One has to make sure that the observed change in behavior could not have been produced by prior separate presentations of the CS or the US. o If exposure to just the US produces increased responding to a previously ineffective stimulus, this is called pseudo-conditioning. Control procedures are required to determine whether responses that develop to a CS represent a genuine CS-US association or just pseudoconditioning. o The control procedure must have the same number and distribution of CS and US presentations as the experimental procedure, but the CS and US arranged so they don’t become associated. Random Control Procedure: present the US periodically during both the CS and the intertrial interval, making sure the probability of the US is the same during the intertrial interval as it is during the CS. There is evidence that this does not prevent conditioned responding. o A more successful control procedure involves presenting the conditioned and unconditioned stimuli on separate trials, which is known as the Explicitly Unpaired Control.
Effectiveness of Common Conditioning Procedures -
-
-
Generally little conditioned responding was observed in simultaneous conditioning procedures, where the CS-US interval (interstimulus interval) was zero. o Delaying the presentation of the US just a little bit after the CS often facilitated conditioned responding. o If it is increased further the conditioned responding declines. Conditioning is most effective when the CS is a good signal for the impending delivery of the US. o The signal value of the CS is the best in the short-delayed procedure, where the US occurs shortly after the onset of the CS. o The CS becomes a less effective signal for the impending delivery of the US as the CS-US interval is increased. o CS also a poor predictor of the US in simultaneous and trace procedures. o Backwards conditioning produced mixed results. The Temporal Coding Hypothesis: the idea that Pavlovian conditioning procedures lead not only to learning that the US happens but exactly when it
occurs in relation to the CS. The CS comes to represent (or code) the timing of the US. Inhibitory Pavlovian Conditioning -
Inhibitory Conditioning is when you learn to predict the absence of the US. The ability to predict bad things is very helpful b/c it also enables you to predict when bad things will not happen. A conditioned inhibitor is a signal for the absence of the US.
Procedures for Inhibitory Conditioning -
Conditioned inhibition has an important prerequisite. o For the absence of a US to be a significant event, the US has to occur periodically in the situation. General Rule: inhibitory conditioning and inhibitory control of behavior occur only if there is an excitatory context for the US in question. o I.e. going to the gas station to fill up on gas, and noticing a sign that says “out of gas”.
Pavlov’s Procedure for Conditioned Inhibition -
The procedure he used involves two conditioned stimuli and two kinds of conditioning trials, one for excitatory and one for inhibitory conditioning. o The US is presented on excitatory conditioning trials and whenever the US occurs, it is announced by a stimulus labeled CS+ (i.e. a tone). B/c of its pairings with the US, the CS+ becomes a signal for the US and can then provide the excitatory context for the development of conditioned inhibition. o During inhibitory conditioning trials the CS+ is presented together with the second stimulus called the CS- (i.e. a light), and the US does not occur. Thus the CS- is presented in the excitatory context provided by the CS+ but the CS- is not paired with the US. This makes the CS- a conditioned inhibitor.
Negative CS-US Contingency or Correlation -
Involves just a CS- that is negatively correlated with the US. o This negative correlation or contingency means that the US is less likely to occur after the CS than at other times. B/c the US occurs periodically in the experimental situation, the contextual cues of the experimental chamber acquire excitatory properties. This then permits the acquisition of inhibitory properties by the CS.
-
In a negative CS-US contingency procedure, the aversive US may occur shortly after the CS occasionally but is much more likely to occur in the absence of the CS. o This contrasts with Pavlov’s procedure where the US always occurs at the end of the CS+ and does not occur when the CS- is presented together with the CS+.
Measuring Conditioned Inhibition Bi-Directional Response Systems -
-
Identification of opposing response tendencies is easy with response systems that can change in opposite directions from baseline or normal performance. o I.e. Heart rate, respiration, and temperature. o I.e. animals can either approach or withdraw from a stimulus or drink more or less of a flavored water. Many responses are not bi-directional. o I.e. A conditioned excitatory stimulus will elicit freezing, but a conditioned inhibitor will not produce activity above normal levels. o A similar problem arises with the eye blink conditioning.
The Compound-Stimulus, or Summation, Test -
-
The compound-stimulus (summation) test is based on the simple idea that conditioned inhibition counteracts or inhibits conditioned excitation. o Therefore to measure how the presentation of a CS- disrupts or suppresses responding that would normally be elicited by a CS+. This test indicates that the presentation of a conditioned inhibitor or safety signal can reduce the stressful effects of an aversive experience. For more see page 94
The Retardation of Acquisition -
The retardation of acquisition test is based on if the stimulus actively inhibits a particular response, then it should be especially difficult to condition that stimulus to elicit the behavior. o In other words, the rate of excitatory conditioning should be retarded if the CS is a conditioned inhibitor. o For more see page 96
Studies of classical conditioning began with the work of Russian physiologist Pavlov. o Started when he noticed that dogs created psychic secretions every time they were presented food. Discoveries of Vul’fson and Snarskii -
-
-
Performed the first systematic studies of classical conditioning. Vul’fson focused on salivary glands. o Studied salivary responses to various substances placed in the mouth: dry food, wet food, sour water and sand for example. o If they had the food placed in their mouth repeatedly, the mere sight of these foods was enough to make them salivate. Snarskii extended the studies with the use of artificial substances. o Used different colors to color bowls of water. Some substances produced distinctive taste and texture sensations in the mouth, known as orosensory stimuli. First time only the feeling of sand in dogs mouth elicited salivation, however after the sand had been placed in the mouth several times the sight of sand also came to elicit salivation. o The association of one feature of an object with another is called object learning. To study the mechanisms of associative learning, the stimuli to be associated have to be manipulated independently of one another. o Pavlov used procedures in which the stimuli to be associated came from different sources.
The Classical Conditioning Paradigm -
Involves two stimuli: one of these stimuli is a tone or a light. The other stimulus is food or the taste of sour solution placed in the mouth. In contrast to the light or tone, the food or sour taste elicits vigorous salivation even the first time it is presented. o The light or tone referred to as conditional stimulus b/c the effectiveness of this stimulus in eliciting salivation depended pairing it several times with the presentation of food. o The food or sour taste was called the unconditioned stimulus b/c its effectiveness in eliciting salivation did not depend on any prior training. o The salivation that eventually came to be elicited by the tone or light was called the conditioned response, and the salivation that was
always elicited by the food or sour taste was called the unconditioned response. Experimental Situations -
Pavlov used dogs using salivary-fistula technique. Most contemporary experiments on Pavlovian conditioning are carried out with rats, rabbits, and pigeons.
Fear Conditioning -
Little Albert Experiment o Conditioned rat (CS) with loud abstract noise (US) and made little Albert afraid (CR) of the rat. When rats are scared they become immobile. Two different indirect measures of immobility known as conditioned suppression procedures are: o Lick Suppression Procedure: rat licking a spout of water, then presented with a fear CS (i.e. tone). Now measure how long it takes to lick a certain amount of times. o Conditioned Emotional Response: rats are first trained to press a response lever for food reward in a small chamber. Once the rats are pressing the lever at a steady rate fear conditioning is introduced consisting of a tone or light paired with a brief shock. As they acquire the conditioned fear, they come to suppress their level pressing during the CS. They calculate a ratio and compare to the normal baseline.
Eye Blink Conditioning -
Puff of air to the eye (US) was given when a tone was heard (CS). Eventually began to blink when they heard the tone.
Sign Tracking -
One experimental paradigm that has contributed significantly to modern conceptions of Pavlovian conditioning is the sign tracking (autoshaping). o Not a highly reflexive response like salvation and blinking. Sign tracking is investigated by presenting a discrete, localized visual stimulus just before each delivery of a small amount of food. o I.e. Pigeon in a chamber with a small circular key that could be illuminated and that the pigeons could peck. Periodically the pigeons were given access to the food for a few seconds. The key light was illuminated for 8 seconds immediately before each food delivery. o Instead of using the key light to tell them when they should go the food dish, they started pecking the key itself.
-
o Interesting behavior discovery b/c the key was not needed to gain access to the food. For sign tracking to occur the CS has to be of the proper modality and configuration.
Learning What Tastes Good or Bad -
-
-
-
A conditioned taste aversion is learned if ingestion of a novel flavor is followed by an aversive consequence such as indigestion or food poisoning. In contrast a taste preference may be learned if a flavor is paired with nutritional repletion or other positive consequences. o This can happen in just one trial and the learning can even occur if the illness is delayed several hours after ingestion of the food. In 20% of the cases, individuals believed that their illness was not the result of the food they ate. This indicates that food aversion learning can be independent of rational thought processes and can go against a person’s conclusions about the causes of their illness. I.e. eating before chemotherapy, most people think their loss of appetite is caused by the chemo, but it could be that eating before produced a taste aversion. Taste aversion learning is a result of the pairing of a CS (taste) and a US (drug injection or radiation exposure) in much the same manner as in other examples of classical conditioning. However it does have some special features: o 1. Strong taste aversions can be learned with just one pairing of the flavor and the illness. o 2. Taste aversion learning occurs even of the illness does not occur until several hours after exposure to the novel taste. Evaluative Conditioning: when our evaluation or liking of a stimulus changes by virtue of having that stimulus associated with something we already like or dislike. o Used extensively in the advertisement industry.
Excitatory Pavlovian Conditioning Procedures -
So far we have been learning excitatory Pavlovian conditioning. In excitatory conditioning, organisms learn an association b/w the conditioned and unconditioned stimuli. o As a result of the association, presentation of the CS activates behavioral and neural activity related to the US in the absence of the actual presentation of the US.
Common Pavlovian Conditioning Procedures -
Major factor that determines the course of classical conditioning is the relative timing of the CS and US.
-
Five Common Classical Conditioning Procedures: o Short-delayed Conditioning: Most frequently used. Involves delaying the start of the US slightly after the start of the CS on each trial. Critical feature is that the CS starts each trial and the US is presented after a brief delay. The CS may continue during the US or end when the US begins. o Trace Conditioning: Similar to short delayed procedure. However in trace conditioning, the US is not presented until some time after the CS has ended. This leaves a gap b/w CS and US called the trace interval. o Long-delayed Conditioning: In this case the US is delayed much longer (5-10 mins) than in the short delayed conditioning. Does not include a trace interval. The CS lasts until the US begins. o Simultaneous Conditioning: Presents the CS and US at the same time. o Backward Conditioning: US occurs shortly before, rather than after, the CS.
Measuring Conditioned Responses -
-
To make comparisons among the various procedures, one has to use a method for measuring conditioning that is equally applicable to all the procedures. o This is typically done with the use of a test trial, which consists of presenting the CS by itself. Responses elicited by the CS can then be observed without contamination from responses elicited by the US. Behavior during the CS can be quantified in several ways. o One aspect of conditioned behavior is how much of it occurs. This is called the magnitude of the CR. I.e. Pavlov measured the amount of saliva that was elicited by the CS. o The energy of responding can also be measured by how often the CS elicits a CR. I.e. measure the percentage of trials on which a CR is elicited by the CS. This reflects the likelihood, or probability of responding. o A third aspect of conditioned responding is how soon the CR occurs after the presentation of the CS. This is known, as the latency of the CR. Latency is the amount of time that elapses b/w the start of the CS and the occurrence of the CR.
Control Procedures for Classical Conditioning -
To be certain that a conditioning procedure is responsible for certain changes in behavior; those changes must be compared to the effects of a control procedure. What should the control procedure be? o One has to make sure that the observed change in behavior could not have been produced by prior separate presentations of the CS or the US. o If exposure to just the US produces increased responding to a previously ineffective stimulus, this is called pseudo-conditioning. Control procedures are required to determine whether responses that develop to a CS represent a genuine CS-US association or just pseudoconditioning. o The control procedure must have the same number and distribution of CS and US presentations as the experimental procedure, but the CS and US arranged so they don’t become associated. Random Control Procedure: present the US periodically during both the CS and the intertrial interval, making sure the probability of the US is the same during the intertrial interval as it is during the CS. There is evidence that this does not prevent conditioned responding. o A more successful control procedure involves presenting the conditioned and unconditioned stimuli on separate trials, which is known as the Explicitly Unpaired Control.
Effectiveness of Common Conditioning Procedures -
-
-
Generally little conditioned responding was observed in simultaneous conditioning procedures, where the CS-US interval (interstimulus interval) was zero. o Delaying the presentation of the US just a little bit after the CS often facilitated conditioned responding. o If it is increased further the conditioned responding declines. Conditioning is most effective when the CS is a good signal for the impending delivery of the US. o The signal value of the CS is the best in the short-delayed procedure, where the US occurs shortly after the onset of the CS. o The CS becomes a less effective signal for the impending delivery of the US as the CS-US interval is increased. o CS also a poor predictor of the US in simultaneous and trace procedures. o Backwards conditioning produced mixed results. The Temporal Coding Hypothesis: the idea that Pavlovian conditioning procedures lead not only to learning that the US happens but exactly when it
occurs in relation to the CS. The CS comes to represent (or code) the timing of the US. Inhibitory Pavlovian Conditioning -
Inhibitory Conditioning is when you learn to predict the absence of the US. The ability to predict bad things is very helpful b/c it also enables you to predict when bad things will not happen. A conditioned inhibitor is a signal for the absence of the US.
Procedures for Inhibitory Conditioning -
Conditioned inhibition has an important prerequisite. o For the absence of a US to be a significant event, the US has to occur periodically in the situation. General Rule: inhibitory conditioning and inhibitory control of behavior occur only if there is an excitatory context for the US in question. o I.e. going to the gas station to fill up on gas, and noticing a sign that says “out of gas”.
Pavlov’s Procedure for Conditioned Inhibition -
The procedure he used involves two conditioned stimuli and two kinds of conditioning trials, one for excitatory and one for inhibitory conditioning. o The US is presented on excitatory conditioning trials and whenever the US occurs, it is announced by a stimulus labeled CS+ (i.e. a tone). B/c of its pairings with the US, the CS+ becomes a signal for the US and can then provide the excitatory context for the development of conditioned inhibition. o During inhibitory conditioning trials the CS+ is presented together with the second stimulus called the CS- (i.e. a light), and the US does not occur. Thus the CS- is presented in the excitatory context provided by the CS+ but the CS- is not paired with the US. This makes the CS- a conditioned inhibitor.
Negative CS-US Contingency or Correlation -
Involves just a CS- that is negatively correlated with the US. o This negative correlation or contingency means that the US is less likely to occur after the CS than at other times. B/c the US occurs periodically in the experimental situation, the contextual cues of the experimental chamber acquire excitatory properties. This then permits the acquisition of inhibitory properties by the CS.
-
In a negative CS-US contingency procedure, the aversive US may occur shortly after the CS occasionally but is much more likely to occur in the absence of the CS. o This contrasts with Pavlov’s procedure where the US always occurs at the end of the CS+ and does not occur when the CS- is presented together with the CS+.
Measuring Conditioned Inhibition Bi-Directional Response Systems -
-
Identification of opposing response tendencies is easy with response systems that can change in opposite directions from baseline or normal performance. o I.e. Heart rate, respiration, and temperature. o I.e. animals can either approach or withdraw from a stimulus or drink more or less of a flavored water. Many responses are not bi-directional. o I.e. A conditioned excitatory stimulus will elicit freezing, but a conditioned inhibitor will not produce activity above normal levels. o A similar problem arises with the eye blink conditioning.
The Compound-Stimulus, or Summation, Test -
-
The compound-stimulus (summation) test is based on the simple idea that conditioned inhibition counteracts or inhibits conditioned excitation. o Therefore to measure how the presentation of a CS- disrupts or suppresses responding that would normally be elicited by a CS+. This test indicates that the presentation of a conditioned inhibitor or safety signal can reduce the stressful effects of an aversive experience. For more see page 94
The Retardation of Acquisition -
The retardation of acquisition test is based on if the stimulus actively inhibits a particular response, then it should be especially difficult to condition that stimulus to elicit the behavior. o In other words, the rate of excitatory conditioning should be retarded if the CS is a conditioned inhibitor. o For more see page 96
