Pavlovian Conditioning: Key Concepts and Information Content

Pavlovian conditioned stimuli – what do they do and what information do they contain

• Pavlovian conditioned stimuli (CS) signal information about likely events (most importantly, the unconditioned stimulus, US).
• They carry predictive information about the occurrence of the US and thereby influence behavior.
• They can elicit approach and consummatory responses when they signal appetitive events (e.g., food, drink, sex) and drug-related effects.
• They can be arbitrary or complex stimuli (e.g., logos, brands) that acquire predictive value through conditioning.
• Examples from everyday life shown in slides: Coca-Cola, BOOST, Juice bars, SUBWAY, McDonald’s, Mars, KFC, Neste, etc. These illustrate that conditioned stimuli can be associated with rewards across contexts.
• CSs convey information about the commodities they signal (not just a simple excitation): they tell the organism what to expect and guide behavior accordingly.
• Sign-tracking is a key phenomenon: organisms approach the CS itself (the sign) rather than the site of the reward; this reflects the information value of the cue as a predictor.
• CSs can acquire appetitive properties (signs that signal food, water, mates) and elicit approach and consummatory responses in pigeons, rats, and humans.
• In drug-related contexts, stimuli signaling drug effects elicit drug-related responses in humans.
• The content of learning via the CS is not just a link between a neutral event and a reflex; it includes the value and nature of the predicted outcome.

The study of Pavlovian conditioning – key questions and aims

• Pavlov brought learning into the laboratory for precision and control.
• The major questions addressed in Pavlovian conditioning are:
  1) When does learning occur?
  2) How is learning expressed in behaviour?
  3) What is it that is learned?
• These questions guide how researchers interpret CS–US pairings, conditioning strength, and the form of conditioned responses (CRs).

Relationship to instrumental (Skinnerian) conditioning

• Skinner focused on how behaviour changes the environment and is in turn shaped by the environment.
• Reinforcement schedules influence persistence and response characteristics.
• Partial reinforcement increases persistence of behaviour; different reinforcement schedules yield different response patterns (e.g., persistence, post-reinforcement pauses).
• Types of reinforcement/punishment and omission training:
  • Positive reinforcement: a response (R) produces a desirable outcome (O).
  • Negative reinforcement: a response prevents an aversive outcome.
  • Punishment: a response produces an aversive outcome.
  • Omission training: a response prevents a desired outcome.

Reinforcement schedules (Skinner) – key points

• Ratio schedules produce rapid rates of responding.
• Fixed ratio (FR) schedules: post-reinforcement pauses; the pause length is proportional to the number of responses required.
• Variable ratio (VR) schedules: produce very constant rates of responding; little or no post-reinforcement pause.
• Interval schedules:
  • Fixed interval (FI): reinforcement is contingent on a response after a fixed time; studying before an exam would fall under FI-like structure; easy-to-spot post-reinforcement pauses and re-initiation before next opportunity.
  • Variable interval (VI): reinforcement at unpredictable times; fosters steady but moderate responding; e.g., checking emails yields a relatively constant checking behavior.

Examples of interval schedules (illustrative)

• Fixed interval: exams occur at regular intervals; studying tends to occur shortly before the exam and then pause afterward.
• Variable interval: checking emails/messages yields a relatively constant rate of checking because reinforcement is unpredictable.

Instrumental conditioning procedures – concise map

• Reinforcement and punishment schemas in instrumental conditioning:
  • Positive reinforcement: R → O (desirable outcome follows the behavior).
  • Negative reinforcement: R prevents O (removal of an aversive outcome following the behavior).
  • Positive punishment: R → aversive outcome (undesirable outcome follows the behavior).
  • Omission training: Behavior prevents a desired outcome.

Sign-tracking and the power of the sign

• Sign-tracking vs. goal-tracking: individual differences in how predictive cues are approached and interacted with.
• Attractive signals draw approach and contact; CSs that signal appetitive events elicit approach and consummatory responses.
• Sign-tracking responses persist even when the CS prevents the US (omission); this indicates strong cue salience and automatic attribution of incentive value to the CS.
• Pigeons show sign-tracking with a food signal (e.g., a red dot) that, when pecking is ceased, and grain is no longer delivered, the CS presentation still elicits conditioned approach behaviors; pecking at the CS can be reinstated under certain conditions (e.g., after extinction followed by CS presentation).
• Williams & Williams (1969) showed sign-tracking persisted even when pecking caused omission of food; pecking eventually ceased only when grain was withheld altogether.
• Le Pelley et al. (2015) extended this to humans: participants could not inhibit looking at a colour signal that predicted a large reward if looking away caused omission of the reward; failure to inhibit gaze showed analogous sign-tracking behavior in humans.

Sign-tracking vs goal-tracking in rats – relevance to addiction

• Individual differences: some rats are sign-trackers, others goal-trackers.
• Sign-trackers are more likely to develop drug-related habits; cues acquire incentive salience and become highly compelling (i.e., want-and-lust value) rather than merely predicting the outcome.
• Compared to goal-trackers, sign-trackers assign greater incentive salience to cue and drug-related cues, increasing vulnerability to addiction.

Preparation: Pavlovian conditioning and evolutionary function

• Pavlovian conditioning contributes to reproductive fitness and defence across species.
• Examples of preparation effects:
  • Male quail: a place that signals females leads to increased copulatory success and offspring, indicating derived fitness benefits from cue learning (Domjan, Psychological Science, 2007).
  • Male blue gourami fish: a light signaling a rival increases territorial defense success and reproductive potential when the rival is introduced after cue presence.
• Preparation Pavlovian conditioning promotes defence by enabling organisms to anticipate danger and respond with sympathetic arousal, reduced pain sensitivity, protective reflexes, and “fight or flight” responses.

Threat or fear conditioning in rats

• Typical CS–US pairing in fear conditioning: context or discrete cues (CS) paired with an aversive US (e.g., foot shock) to elicit conditioned fear (CR).
• Classic setup: context (CS) paired with shock (US) leading to CR such as freezing.

Content of Pavlovian conditioning is shaped by biology

• The cue-to-consequence effect (Garcia & Koelling, 1966): some CS–US pairings are learned more readily due to evolutionary history.
• Experimental setup: flavoured water (CS) paired with US that varies by group:
  • Group 1: flavoured water paired with shock after drinking.
  • Group 2: flavoured water paired with sickness from X rays while drinking.
• Later, rats were tested with flavoured water versus a different cue (bright-noisy water).
• Findings:
  • Group 1 learned the association between the noise (a novel cue) and shock, but did not generalize a shock–noise link to flavour; they did not form a strong flavour–shock link.
  • Group 2 learned the association between flavour and sickness, but not between the noise and shock.
• Conclusion: Learners show selective associations that reflect adaptive biases: mice and rats form taste aversions (nausea) more readily than fear associations with unrelated cues; this is known as the Garcia effect.

Sign-tracking and addiction – summary of biology-informed learning

• Pavlovian conditioned responses are powerful and reveal the content of what is learned.
• Across species, many conditioned responses are difficult or nearly impossible to suppress because of their strong biological basis and cue salience.
• Conditioning is shaped by biology; learning helps organisms solve core problems of existence (food, reproduction, defence).

Contiguity vs information in learning – temporal contiguity is not sufficient

• Temporal contiguity (events occurring together in time) is a historic idea about learning: closer timing increases learning likelihood.
• Is contiguity necessary? No: learning can occur with delayed relationships between events; and learning can occur about relations that have never occurred together.
• Is contiguity sufficient? No: even with contiguity, learning does not guarantee that a CS will be informative about the US.

Blocking effect (Kamin, 1969)

• Setup: Stage 1 – a light predicts a shock (CR to the light). Stage 2 – light + sound are presented together, still followed by shock. Test: sound alone is weak or fails to elicit CR.
• Conclusion: Contiguity is not sufficient for learning; the CS must be informative about the US.
• Implication: If a better predictor exists, learning shifts to that predictor and the previously trained CS loses associative strength.

Contingency effect (Rescorla, 1969)

• Setup: The degree to which the CS provides information about the occurrence of the US affects learning.
• Test: Rats learn about the light when it provides reliable information about the shock.
• Result: Contingency, i.e., reliable informational content, drives learning beyond simple contiguity.

Implications of contiguity and contingency in learning

• Temporal contiguity alone does not guarantee associative learning.
• The CS must be informative about the US; learning is guided by predictions and error signals.
• If a better predictor of the US is present (blocked by a superior predictor, or context provides better prediction), learning about the original CS fails or is greatly reduced.
• Learning is driven by prediction errors: when the US is surprising, learning about the CS is strong; as the CS predicts the US more accurately, the surprise diminishes and learning ceases when the CS fully predicts the US.

What is learned in Pavlovian conditioning – recap

• The content of learning includes specific associations (CS–US link) and their predictive value for upcoming events.
• The learned information shapes behavioural responses including approach, avoidance, and defensive reactions.
• The power and content of CS–US learning are strongly modulated by biology, environment, and the predictive usefulness of the cue.

Connections to broader themes and real-world relevance

• Conditioning explains how environmental cues (advertising logos, brands) can acquire motivational value and influence consumption and behavior.
• Sign-tracking and cue salience help explain vulnerability to addiction and the role of environmental cues in relapse.
• Preparation and defensive conditioning illustrate how learning adapts organisms to ecological demands (reproduction, territory defense, danger detection).
• Temporal contiguity and information content provide a framework for understanding why some associations are easily learned while others are not, guiding strategies in education, therapy, and behavior modification.

Summary of core concepts

• CSs are predictors that carry information about the US and guide behavior through learned associations.
• The strength and form of conditioning depend on contiguity, contingency, and informational value of the CS.
• Different types of conditioning and learning (taste aversion, flavour-nutrient learning, sign-tracking, fear conditioning) illustrate the broad applicability of Pavlovian principles across species and contexts.
• Learning is constrained and shaped by biology, which influences which associations are readily formed and which cues gain motivational significance.
• Blocking and contingency demonstrate that learning depends on predictive information, not just temporal proximity.

Key terms to know

• CS: conditioned stimulus
• US: unconditioned stimulus
• CR: conditioned response
• Sign-tracking: approach to the CS itself
• Goal-tracking: approach to the location where the US is delivered
• Contiguity: temporal proximity of CS and US
• Contingency: informational linkage between CS and US
• Garcia effect: biology-guided selectivity in CS–US associations (taste aversion learning vs other sensory modalities)
• Blocking: learning to a previously trained CS reduces learning to a new CS when both predict the same US
• Contingency effect: learning depends on the predictive value of a CS relative to other cues
• Omission training: behavior prevents the occurrence of the desired outcome
• Preparation: evolutionary advantages conferred by conditioning (reproduction, defence)