Classical Conditioning Foundations

Classical Conditioning Foundations

Pavlovian (Classical) Conditioning

• Definition: Primarily, it's the most extensively studied example of associative learning.
• Core Concept: An organism learns to associate a previously neutral stimulus (Conditioned Stimulus, CS) with a biologically significant stimulus (Unconditioned Stimulus, US), leading to an anticipatory response (Conditioned Response, CR) before the US occurs.
• Example (Pavlov's Dogs): A dog learns that the ringing of a "bell" (CS) is associated with the imminent delivery of food (US), causing the dog to prepare to eat by salivating (CR) before the food arrives.

Appetitive Conditioning

• Basic Pairing:
  • First Few Trials: A neutral CS (e.g., bell) is presented, followed by a US (e.g., food), which naturally elicits an Unconditioned Response (UR, e.g., salivation).
  • Later Trials: After repeated pairings, the CS alone elicits a CR (e.g., salivation), which is similar to the UR but is evoked by the CS.
• Notation:
  • CS = Bell
  • US = Food
  • UR = Salivation
  • CR = Salivation

Two Possibilities for How Learning Occurs

1. S-R Learning (Stimulus-Response): The CS (e.g., bell) directly calls up the CR (e.g., salivation). This is a direct association between the CS and the CR.
2. S-S Learning (Stimulus-Stimulus): The CS (e.g., bell) indirectly calls up the CR by signaling the US (e.g., food), which then triggers the CR. This implies an association between the CS and the US representation in the brain.

Excitatory Conditioning

• Mechanism: According to Pavlov, when a CS and US are associated, the CS representation in the brain excites the US representation. This forms an excitatory CS-US association.
• Behavioral Outcome: The organism becomes excited and may approach the CS if possible.
• Example: Dogs have been observed to lick light CSs that predict food, demonstrating an approach toward the excitatory CS.

Central Players: CS and US

• Unconditioned Stimulus (US):
  • Initially "potent": It naturally elicits both hidden and observable changes.
  • Example (Food US):
    • Motivational State (inside and hidden): Reduces hunger.
    • Salivation (outside and observable response): The physical act of salivating.
• Conditioned Stimulus (CS):
  • Initially "neutral": It does not initially trigger the motivational state or observable response related to the US.
  • Examples: Lights, sounds, tastes, odors, etc.
  • After Pairings: After being paired with the US, the CS acquires the ability to elicit conditioned responses (e.g., hunger motivation, salivation).

What are the CRs?

• Source of CRs: The US possesses both sensory and motivational properties that influence the nature of the CRs.
• Consummatory CRs (Sensory Properties of US):
  • Specific to the route of delivery or the physical interaction with the US.
  • Example: Salivation and light bulb licking (if the CS is a light) are consummatory behaviors directed towards the mouth region, suggesting the animal is anticipating the taste or physical sensation of the food.
• Preparatory CRs (Motivational Properties of US):
  • Specific to an internal state, such as hunger or appetitive motivation.
  • Example: Aprroach behavior is a preparatory CR, reflecting the animal's internal motivational state of anticipating a reward.

Pavlovian Aversive Conditioning

• Concept: Conditioning where the US is an unpleasant or aversive stimulus, leading to conditioned fear, withdrawal, or other aversive responses.
• Stefan Vul'fson's Experiment:
  • CS: Sight of natural sand in Vul'fson's hand.
  • Aversive US: Sand placed in the dog's mouth.
  • CR: Dog salivates at the sight of the sand before it's in its mouth, anticipating the unpleasant gritty texture.
• Anton Snarskii's Experiment:
  • CS: Sight of artificial black liquid made sour.
  • Aversive US: Mild acid (the sour liquid) placed in the dog's mouth.
  • CR: Dog learns to salivate at the sight of the black liquid before it's in its mouth, anticipating the sour taste.
• Object Learning: These examples illustrate "object learning," where the animal learns about the properties of an object (sand, sour liquid) that signal an aversive outcome.

Conditioning of Fear in Little Albert

• Experiment: Conducted by John B. Watson and Rosalie Rayner.
• CS: White rat.
• US: Loud gong (unpleasant noise).
• CRs: Little Albert developed fear (crying, crawling away) of the white rat and generalized this fear to other furry stimuli like a rabbit, a dog, and even a Santa Claus mask.
• Signs of Learning (Little Albert):
  • Conditioned Withdrawal / Negative Sign-Tracking: Albert leaned away from the rat, showing a general tendency to move away from aversive stimuli. This suggests he encoded the motivational properties (badness) of the US.
  • Sensory Properties: Evidence that Little Albert encoded the sensory properties of the loud gong itself is less clear.
  • Phobia Connection: This experiment demonstrated that phobias might be learned, though whether Albert truly developed a clinical phobia is questionable.

Aversive Fear Conditioning in Rats

• Setup:
  • CS: Auditory stimulus (e.g., white noise).
  • US: Grid shock applied to the feet.
• Observable CRs (Fear):
  • Freezing (immobility, a primary behavioral measure).
  • Physiological responses: Blood pressure changes, heart rate changes, stress hormone release, analgesia (reduced pain sensitivity).
• Acquisition Curve: Shows the percentage of time rats spend freezing during CS presentation increasing over conditioning trials. (Based on Reger et al., 2012).

Conditioned Suppression (Conditioned Emotional Response, CER)

• Procedure: A common method to measure conditioned fear.
  • Stage 1: The rat learns an operant response, such as bar pressing for a reward (e.g., food pellets) while hungry.
  • Stage 2: A neutral stimulus (CS, e.g., white noise) is paired with an aversive stimulus (US, e.g., grid shock). The lever might or might not be present during this stage.
  • Stage 3: The CS is presented during ongoing lever-press responding.
  • Outcome: Conditioned fear, and specific CRs like freezing, will cause the animal to reduce or stop responding on the lever.
• Suppression Ratio: A quantitative measure of conditioned suppression.
  • Formula: \frac{a}{(a + b)}
    • a = \text{responses during the CS}
    • b = \text{responses prior to the CS}
  • Interpretation:
    • A ratio of 0.50 indicates no suppression (e.g., 30/(30+30)).
    • A ratio closer to 0 indicates strong suppression (e.g., 3/(3+20) = 0.13).
  • Acquisition Curve: Shows the mean suppression ratio decreasing (indicating stronger suppression) over conditioning sessions, often comparing weak vs. strong shock conditions.

Lick Suppression

• Procedure: Another method for measuring conditioned fear, particularly for thirsty animals.
  • Stage 1: A thirsty rat is placed in an experimental box with a water-filled tube.
  • Stage 2: A neutral CS is paired with an aversive US (e.g., shock) without the water tube present.
  • Stage 3: The CS is presented when the rat is placed in the experimental box with the water tube. The time to complete a specified number of licks (latency) is measured.
  • Outcome: Conditioned fear and freezing will cause the animal to avoid licking the spout for a significant period, resulting in longer latencies.

Pavlov's List of Excitatory Procedures

These describe different temporal relationships between the CS and US, leading to varying strengths of CRs:

• Short-Delay Conditioning: CS onset slightly precedes US onset, both overlap. Normally robust CR.
• Trace Conditioning: CS onset, CS offset, then a gap before US onset. CR normally first appears during the CS-US gap and then later during the CS and gap. Weaker CR than short-delay.
• Long-Delay Conditioning: CS onset, then a long delay before US onset, both overlap. Weaker CR than short-delay.
• Simultaneous Conditioning: CS and US presented at the exact same time. Produces little responding, but learning can still occur.
• Backward Conditioning: US onset precedes CS onset. Few trials might lead to a small CR, but more trials typically do not increase it significantly.

Conclusions about Conditioned Responses

• Both specific responses based on the exact nature of the US (its sensory properties) and motivational states (appetitive/aversive properties) are conditioned.
• CSs become associated with USs, resulting in anticipatory conditioned responses that occur prior to the delivery of the US.

Measuring Conditioned Responses

To compare groups effectively, test trials with the CS alone are often used. Typical measures include:

• Magnitude of CR: How much of the response occurs (e.g., Pavlov measured the number of drops of saliva).
• Probability of CR: How likely the CR is to occur (most common measure).
• Latency of CR: How soon the CR occurs after the onset of the CS. This can indicate if the subject has learned when the US is going to be delivered.

Control Procedures for Classical Conditioning

To confirm that observed behavioral changes are due to an established association (and not just prior exposure), control procedures are crucial:

• Pseudo-Conditioning: Occurs if exposure to only the US produces an increased response to a previously ineffective CS, even without CS-US pairings.
• Random Control Procedure: Presents the US at random times during both the CS and the intertrial interval, ensuring the probability of the US is the same during both periods. This prevents the formation of a predictive association.
• Explicitly Unpaired Control: Presents the CS and US on separate trials, ensuring they never co-occur. This demonstrates that contiguity (close temporal pairing) is necessary for conditioning.

Three "Preparations" (Situations) and a Question

• These are laboratory situations commonly studied.
• Key Question: Identify the CS, US, UR, and CR in each preparation.
• Definitions Revisited:
  • CS and US: Controlled by the experimenter.
  • US: A biologically significant event that elicits a behavior without prior pairings.
  • URs: The responses naturally evoked by the US.
  • CRs: Changes in behavior evoked by the CS, resulting from conditioning.

Preparation 1: Thirsty Rabbits and Jaw Movements

• Scenario: Thirsty rabbits increase their jawing motions to a tone that signals water delivery.
• Analysis:
  • CS: Tone.
  • US: Water delivery.
  • UR: Natural jaw movements/ingestive behaviors to water.
  • CR: Increased jawing motions to the tone.
• "Real" Conditioning Evidence: Graphs show higher percentage of responses for the "Group T-W" (paired CS-US) compared to "Group T/W" (mere exposure to CS and US, unpaired) across acquisition blocks, indicating true conditioning.

Preparation 2: Infant Eye-Blink

• Scenario: Infants expecting a gentle puff of air aimed at the right eye blink in advance.
• Analysis:
  • CS: Tone, light, or other neutral stimulus.
  • US: Gentle puff of air to the eye.
  • UR: Natural eye-blink to the air puff.
  • CR: Eye-blink to the CS.
• "Real" Conditioning Evidence: Changes are mostly real, specifically caused by pairings. Control groups are essential to rule out increased responding due to mere exposure to CS, US, or both (pseudo-conditioning).
• Pseudo-conditioning: Occurs when exposure to the US alone increases responding to the CS, even without CS-US pairings.

Preparation 3: Pigeon Autoshaping / Sign-Tracking

• Scenario: A pigeon moves toward and pecks a light that signals the presentation of food.
• Analysis:
  • CS: Light (keylight).
  • US: Food presentation.
  • UR: Natural pecking/consummatory behavior towards food.
  • CR: Movement toward and pecking of the light (conditioned approach or sign-tracking).
• Autoshaping: The pecking of the light, specifically, is sometimes called "autoshaping."

Form of the Keylight Pecking CR

• Specificity: The form of the pecking CR depends on the US, indicating the animal encodes the sensory properties of the pleasant US.
  • US = Water: The pigeon pecks with a closed beak, characteristic of drinking.
  • US = Food (Grain): The pigeon pecks with an open beak, characteristic of eating.

Sign-Tracking

• Definition: Conditioned approach or interaction with the CS itself.
• Examples:
  • Pigeons pecking a light (CS) that predicts food (US).
  • Squirrels predicting acorns based on tree leaves/shape.
  • Male quail approaching a light (CS) that predicts the availability of a female (US).
• Long-Box Sign-Tracking:
  • Male quails may sign-track to a light 8 feet away from the female's lair, even if it's maladaptive (they miss mating opportunities).
  • This highlights a potential "mismatch" between the inflexible Pavlovian CRs and artificial experimental conditions, especially with low spatial contiguity between CS and US.
• Conditions Favoring Sign-Tracking in Long Box: More likely when the CS-US interval is relatively short (e.g., 1 min) rather than long (e.g., 20 min).

Sign-Tracking and Goal-Tracking

• Distinction: When a localized CS and a "pleasurable" US are paired, the CR can be directed at the CS (sign-tracking) or at the location of the US (goal-tracking).
• Goal-tracking: Conditioned behavior that tracks where the goal object (US) will be presented.
  • Example: An inserted lever (CS) predicts food (US). CR: The rat checks the food magazine (US location).
• Sign-tracking: Conditioned behavior directed at the CS itself.
  • Example: An inserted lever (CS) predicts food (US). CR: The rat licks the lever (CS).
• Influencing Factors: The specific situation, species, and experimental procedures can determine whether an animal primarily engages in sign-tracking or goal-tracking.
• Individual Differences: There's considerable variation among animals in developing and exhibiting sign-tracking.
• Correlations: Individual differences in sign-tracking are correlated with impulsivity and vulnerability to drug abuse in rats.

Excitatory and Inhibitory Conditioning

• Excitors (CS+s): Signal that the US is forthcoming. An excitatory CS-US association is acquired.
• Inhibitors (CS-s): Signal that an otherwise likely US will not occur. A CS-no US association is acquired.
  • Distinction: An inhibitor actively signals non-occurrence; it's not merely a neutral stimulus.

Information Matrix

This matrix illustrates what CS+ and CS- signal for different US types:

• A CS+ prepares the animal for a forthcoming US; a CS- tells the animal the US is not forthcoming.

Pavlov's A+/AB- Procedure (for Inhibitory Conditioning)

• Procedure:
  1. A+: Stimulus A (e.g., bell) is alone, followed by the US (e.g., food). This establishes A as an excitor.
  2. AB-: Stimuli A and B (e.g., bell + light) are presented together, followed by no US. This conditions B as an inhibitor because B predicts the absence of the US when A would usually predict its presence.

Measuring Conditioned Inhibition

• Challenge: After A+, AB- training, B alone does not evoke a CR. It appears neutral, making direct measurement difficult. Skinner argued that AB might be treated as a new configuration, and true inhibition might not exist.

Solution #1: Bidirectional Baseline

• Concept: Some baselines allow for responses both above and below a neutral point, making inhibition observable.
• Example (Pigeon Sign-Tracking):
  • A hungry pigeon approaches a CS+ (signals food).
  • If a CS- is encountered (signals no food), the pigeon will actively withdraw from the CS-.
  • Measuring floor tilt (e.g., towards grain hopper for high score, towards rear for low score) provides a quantitative measure of approach vs. withdrawal behavior, reflecting excitation vs. inhibition.

Solution #2: Two Test Strategy

To definitively prove that B (from A+, AB- training) is an inhibitor, two types of tests are used:

• 1. Retardation Test:

  • Hypothesis: If a suspected CS- truly signals "no US," it should be slow to become a CS+ if it is later reinforced alone.
  • Procedure (Example for B as CS-):
    • Stage 1 (Inhibitory Training): Group receives A+, AB- training (B becomes a suspected CS-).
    • Test Stage: B is then paired alone with the US (B+) for several trials.
  • Result: If B is an inhibitor, its acquisition as an excitor (i.e., making it a CS+) will be significantly slower than for a control stimulus that received no prior inhibitory training.
  • Graphical Representation: An experimental group (starting below "0" on an excitation-inhibition scale due to inhibition) takes more trials to reach "0" and then positive excitation compared to a control group (starting at "0").
  • Data Example: Shows that a trained inhibitor (CS-) results in less fear (e.g., higher suppression ratio, less lick suppression) when subsequently paired with the US than a control CS.

• 2. Summation Test:

  • Hypothesis: If a suspected CS- signals "no US," it should suppress responding to a different existing CS+ (a "transfer excitor") when presented together.
  • Procedure (Example for B as CS-):
    • Stage 1 (Training): Group receives A+, AB- training (B becomes a suspected CS-), and also establishes another stimulus C as an excitor (C+).
    • Test Trials: Present C alone, and then present C and B together (CB).
  • Result: If B is inhibitory, the CR to the compound stimulus (CB) should be less than the CR to C alone, as B sums with C to reduce the overall excitatory effect.
  • Data Example: Shows that a compound stimulus (CB) elicits less fear (e.g., lower lick suppression score, meaning less suppression) than C alone, demonstrating the inhibitory effect of B.

Other Inhibitory Procedures

Besides Pavlov's A+, AB- procedure, other conditions can produce an inhibitory CS:

• Trace Conditioning with a Long Gap: If there's a long interval between CS offset and US arrival, the CS can sometimes become inhibitory (Pavlov, 1927).
• Many Backward Conditioning Trials: Repeated presentations of the US before the CS, rather than just a few trials, can lead to the CS becoming an inhibitor (Heth, 1976).
• Unpaired CS and US (Negative Correlation/Contingency): When the CS and US are negatively correlated (i.e., when one is present, the other is typically absent), the CS can become inhibitory (Rescorla, 1968).