← all systems · see it in The System
Classical conditioning
Classical (Pavlovian, respondent) conditioning is a type of learning which occurs when a conditioned stimulus (CS) is paired with an unconditioned stimulus (US); after pairing, the CS alone elicits a conditioned response (CR). Modern redefinition: the process whereby experience with a conditional relationship between stimuli bestows these stimuli with the ability to promote adaptive behavior patterns that did not occur before the experience.
Properties published scienceour proposal
Sourced fields, and our own readings.
| born at | kind: probably early Cambrian, about 540 Mya (contested lower); one association: the first pairing |
| born precision | proposed, not dated directly |
| boundary | Inside: the learning process the field names (acquisition, extinction and recovery, generalization and discrimination, blocking, latent and conditioned inhibition, second-order conditioning), from the pairing of a neutral and a biologically potent stimulus, through the sites where their signals converge and the prediction error that sets how much is learned, to the stored association, its gating by context, and the conditioned response it releases. Outside: the senses and reflexes it borrows (the unconditioned reflex is inborn), the memory systems that keep the trace, operant learning (behaviour shaped by its consequences), and the therapies, advertising and schools of thought that run on it. |
| nested level | A kind of associative learning and of nondeclarative memory, run by nervous systems: convergence sites in the amygdala (fear), cerebellar Purkinje cells (eyeblink), midbrain dopamine neurons (reward prediction error), sensory-motor synapses in Aplysia; part of the memory floor of the brain. |
| scale | organism |
Measures (20) published science
| conditioned salivation latency (metronome) | secretion begins after 9 s; 11 drops in 45 s s; drops |
| unconditioned salivation latency (food in mouth) | 1-2 s |
| eyeblink CS-US interval range | 0.1 to 2.0 s |
| trials to the first conditioned blink | a few to several hundred trials |
| Purkinje pause acquisition, fastest cells | trial 2 (after one CS-US pairing), 4 of 10 cells trials |
| inter-trial interval in the decerebrate ferret protocol | 15 s |
| cerebellar cortex microzones | about 1,000 microzones |
| mice learning a 250 ms trace interval | 75 % |
| taste aversion CS-US delay (non-human animals) | 1-6 h strong; reported up to 24 h |
| taste aversion CS-US delay (humans, questionnaires) | up to 7 h |
| phyla with associative learning | at least 6, possibly 9 phyla |
| first limited associative learning (proposed) | about 540 Mya (early Cambrian) |
| UAL in molluscs after arthropods and vertebrates | about 250 million years later |
| pea seedlings following the conditioned fan (Gagliano) | 62% (F+L) and 69% (F vs L) % of seedlings |
| pea replication test (Markel) | p = 0.335, n = 61 + 60 (original p = 0.0027, n = 13 + 10) Fisher's exact test |
| Gelber's Paramecium training dip | 15 s per reinforced trial |
| energetic cost, synaptic versus intracellular computation (claim) | roughly 13 orders of magnitude greater for synaptic ratio |
| hippocampal genes with altered methylation after fear conditioning | 9.2 % of genes (rats) |
| honeybee PER conditioning protocol introduced | 1961 year |
| Pavlov's experimental results published | 1897 year |
Open contentions (5) open contention
Where sources or readings disagree. Both stay on the map.
- C-2520 Can creatures without neurons be conditioned?
booked: The map keeps the competing readings; none is ruled.
the other reading: rung 6 or lower: single cells condition (Gelber's Paramecia, rehabilitated by Gershman et al. 2021); plants condition (Gagliano et al. 2016), not replicated blind (Markel 2020); nematode conditioning may be conditional sensitization (Bhatla via Bronfman) - C-2526 Does extinction erase a learned link, or write over it?
booked: The map keeps the competing readings; none is ruled.
the other reading: extinction unlearns (the Rescorla-Wagner reading drives V back to zero); both, in proportions that differ by protocol - C-2527 Does conditioning need the forebrain?
booked: The map keeps the competing readings; none is ruled.
the other reading: Pavlov 1927: a decerebrate dog would never form the salivary CR; Gallistel et al. 2022 and the decerebrate literature: the brain stem alone suffices for the blink - C-2549 Is classical conditioning a procedure or a process?
booked: The map keeps the competing readings; none is ruled.
the other reading: a behavioral procedure (encyclopedia's first sentence); a process (Fanselow and Wassum's redefinition) - C-2558 Is memory stored in synapses or inside cells?
booked: The map keeps the competing readings; none is ruled.
the other reading: synaptic plasticity is the substrate (prevailing theory); memory is stored intracellularly (Gallistel; Gershman et al.)
Partners and parts published science
Built from / builds (9)
- Synapse
- Cerebellum
- Amygdala
- Prefrontal cortex
- Hippocampus (memory)
- Purkinje cells
- Midbrain dopamine neurons
- Engram (memory trace)
- NMDA receptor
Sits inside / holds (4)
Trades with (valence) (15)
- Cerebellum via sys:electrochemical-gradient · Classical conditioning gain
- Basal ganglia via sys:neuromodulator · Classical conditioning gain
- Amygdala via sys:electrochemical-gradient · Classical conditioning gain
- Human (Homo sapiens) via sys:sensory-environment · Classical conditioning gain
- Prefrontal cortex via sys:electrochemical-gradient · Classical conditioning mixed
- Plant (kingdom Plantae) via sys:electromagnetic-field · Classical conditioning even
- Immune system via sys:blood · Classical conditioning gain
- Emotion (feelings) via sys:sensory-environment · Classical conditioning gain
- Hippocampus (memory) via sys:electrochemical-gradient · Classical conditioning gain
- Digestive system (the gut and its accessory organs) via sys:aqueous-solution · Classical conditioning gain
- Honeybee (western honey bee, Apis mellifera) via sys:sensory-environment · Classical conditioning gain
- Memory via sys:electrochemical-gradient · Classical conditioning gain
- Learning via sys:sensory-environment · Classical conditioning gain
- Dog via sys:sensory-environment · Classical conditioning gain
- Operant conditioning via sys:sensory-environment · Classical conditioning mixed
expressed by (5)
- Uncertainty Reduction/Learning
- Trail-Marking / Externalized Memory (Stigmergy)
- Quick Harm Avoidance
- Proactive Defense
- Technique-Based Identification (a.k.a. Technique-Based Detection)
floor filled by (18)
- Synapse
- Nervous system
- Cerebellum
- Amygdala
- Thalamus
- Prefrontal cortex
- The world through the senses (sensory environment)
- Immune system
- Hippocampus (memory)
- Digestive system (the gut and its accessory organs)
- Memory
- Purkinje cells
- Inferior olive
- Midbrain dopamine neurons
- Engram (memory trace)
- NMDA receptor
- Rescorla-Wagner model
- Experimenter (laboratory procedure)
represents (1)
Kinds (16)
- Fear (threat) conditioning
- Associative learning
- Nondeclarative memory
- Learning
- Eyeblink conditioning
- Food-based appetitive conditioning
- Conditioned taste aversion
- Evaluative conditioning
- Second-order (higher-order) conditioning
- Trace conditioning
- Delay conditioning
- Temporal conditioning
- Backward conditioning
- Conditioned immune and neuroendocrine responses
- Olfactory conditioning of the proboscis extension response (honeybee)
- Extinction (psychology)
Notes published science
- A cerebellar Purkinje cell can learn the gap between a touch and a puff to the eye and store it as one single, unusually long pause between two spikes, timed so the eyelid shuts just as the puff is due; in some cells one pairing was enough (src:gallistel-2022).
- Pavlov wrote that a dog without its cerebral hemispheres would never learn a conditioned reflex, yet the conditioned blink forms in decerebrate cats, rabbits, ferrets and guinea pigs: the brain stem alone suffices (src:pavlov-1927-lecture2; src:gallistel-2022).
- A divorced mother of three who went back to graduate school trained single-celled Paramecia to approach a clean wire that had been baited with bacteria in the 1950s; her critics won and she was forgotten, until a 2021 review argued she had been right (src:gershman-2021).
Layers
The ways it takes in its world, where they meet, and how it acts back, each from L1 (where it meets the world) up.
classical conditioning: from paired events to a learned signal and a prepared response (reader's view)
- L1: Two events in the world in a time relation: a neutral stimulus (tone, light, touch, flavour, place, time) and a biologically potent one (food, shock, air puff, poison)
- L2: The senses take each stimulus in: the ear for the metronome, receptors of mouth and tongue for food; the unconditioned reflex is inborn
- L3: Pathways carry CS and US to where they meet: sensory thalamus to lateral amygdala (fear); mossy fibres (CS) and climbing fibres (US) to the cerebellum (eyeblink); sensory neurons and facilitatory input in Aplysia
- L4: Convergence sites (the ports): lateral and basolateral amygdala neurons with NMDA receptors as coincidence detectors; Purkinje cells; adenylyl cyclase at the Aplysia SN-MN synapse, where the two signals converge
- L5: Prediction error sets how much is learned: the change in associative strength is the gap between what the US supports and what all present cues already predict (Rescorla-Wagner); dopamine neurons signal reward prediction error
- L6: The stored association and its gating: an engram (in Purkinje cells, the CS-US interval); extinction writes a new CS-no US association rather than erasing the first, and context (hippocampus) and prefrontal cortex gate which one is expressed
- L7: The CS becomes a signal for the US and sets off a prepared, timed response back on the world (salivation, a blink timed to the threat, freezing, rejecting a flavour, changed immune and endocrine state, liking)
the boardroom (shared floors)
- L5: Prediction error: learning is driven by the gap between the US and what all present cues predict (Rescorla-Wagner); dopamine neurons signal this gap for reward; error correction regulates learning in both the fear and the appetitive circuit
- L6: The stored association and its gating: the interval engram in Purkinje cells; extinction as a new, context-dependent CS-no US association; prelimbic and infralimbic cortex gate expression; the hippocampus gates by context
- L7: Signalization: the CS now stands for the US and sets off a prepared response, timed to when the US is expected
way in: Context (the place where pairing happens)
- L1: The room and apparatus: Pavlov's dogs salivated as they approached it
- L4: Hippocampus: contextual gating of extinction and context conditioning
way in: Flavour as CS (taste and odour of a food)
- L1: The orosensory properties of a food: odour, taste, temperature, texture
way in: Food as US (appetitive)
- L1: Food in the mouth
- L2: Receptors in the mucous membrane of mouth and tongue
- L3: The inborn salivary reflex: secretion in 1 to 2 seconds
- L4: Midbrain dopamine neurons respond to the reward
way in: Illness as US (visceral malaise after a poison)
- L1: Gastrointestinal malaise, illness, nausea after a food-borne poison
way in: Light or a picture as CS
- L1: A light, a picture
- L4: Dopamine neurons respond to the reward-predicting stimulus itself
way in: Pain as US (footshock, a bite)
- L1: A mild electric shock; the pain of being bitten
- L4: Amygdala subnuclei where the tone-shock association forms
way in: Sound as CS (bell, metronome, tone)
- L1: A sound: a beating metronome, a tone
- L2: The ear: the sound falls upon the ear
- L3: Auditory thalamus to the lateral amygdala
- L4: Lateral and basolateral amygdala neurons receiving thalamic glutamate input
way in: Threat to the eye as US (air puff)
- L1: A puff of air or other threat to the eye
- L3: Climbing fibres from the inferior olive carry the US
- L4: Purkinje cells in the blink-controlling microzone
way in: Time as CS (temporal conditioning)
- L1: A US on a regular schedule: the passage of time itself
- L4: Purkinje cells hold the CS-US interval (the interval engram)
way in: Touch as CS (forepaw, siphon)
- L1: Touch on the skin: stimulation of the dorsum of the forepaw
- L2: Sensory neurons of the skin (in Aplysia, the siphon sensory neurons)
- L3: Mossy fibres carry the CS to the cerebellum (in Aplysia, sensory neuron to motor neuron)
- L4: Purkinje cell via parallel fibres; in Aplysia the SN-MN synapse with adenylyl cyclase
way out: Conditioned fear responses (freezing, startle, autonomic)
- L1: Freezing; a fall in heart rate where shock raised it
- L5: The amygdala central nucleus drives hypothalamus, periaqueductal gray, pons and brainstem
- L7: The tone signals the shock
way out: Conditioned immune and neuroendocrine responses
- L1: Cellular and humoral immune and neuroendocrine functions changed
- L7: A taste paired with an immune-acting drug comes to change immune function
way out: Evaluative conditioning and motivation (liking; Pavlovian-instrumental transfer)
- L1: Attitudes toward products and logos; faster lever pressing in the presence of the sound
- L7: A CS paired with a positive or negative US becomes liked or disliked
way out: Salivation and digestive preparation
- L1: Saliva secreted: from 9 seconds, eleven drops in 45 seconds; the dog turns toward the food place and licks its lips
- L7: The metronome is the signal for food
way out: Taste aversion: palatability downshift and reduced intake
- L1: The flavour tastes worse and less is eaten
- L7: The flavour signals the illness
way out: The conditioned blink
- L1: The eyelid or nictitating membrane closes
- L5: The pause releases the cerebellar nuclei: an excitatory output signal
- L6: The Purkinje cell pause read out from the interval engram
- L7: The blink timed to peak when the threat is anticipated
Sources (22) published science
Schultz W (2016) Dopamine reward prediction error coding. Dialogues Clin Neurosci 18:23-32. doi:10.31887/DCNS.2016.18.1/wschultz (PMC4826767, open) (behind its links)
Hawkins RD, Byrne JH 2015. Associative Learning in Invertebrates. Cold Spring Harb Perspect Biol. doi:10.1101/cshperspect.a021709 (PMC4448622) (behind its links)
Christian KM, Thompson RF 2003. Neural substrates of eyeblink conditioning: acquisition and retention. Learn Mem. doi:10.1101/lm.59603 (behind its links)
Pavlov IP 1927. Conditioned Reflexes, Lecture II (trans. Anrep), Classics in the History of Psychology (behind its links)
Lin JY, Arthurs J, Reilly S 2017. Conditioned taste aversions: From poisons to pain to drugs of abuse. Psychon Bull Rev. (PMC5857059, author manuscript) (behind its links)
Gagliano M, Vyazovskiy VV, Borbely AA, Grimonprez M, Depczynski M 2016. Learning by association in plants. Sci Rep. doi:10.1038/srep38427 (behind its links)
Markel K 2020. Lack of evidence for associative learning in pea plants. eLife. doi:10.7554/eLife.57614 (behind its links)
Gershman SJ, Balbi PE, Gallistel CR, Gunawardena J 2021. Reconsidering the evidence for learning in single cells. eLife. doi:10.7554/eLife.61907 (behind its links)
Bouton ME 2004. Context and behavioral processes in extinction. Learn Mem. doi:10.1101/lm.78804 (behind its links)
Learning (Wikipedia, via Wikimedia Enterprise) (behind its links)
Operant conditioning (Wikipedia, via Wikimedia Enterprise snapshot) (behind its links)