Section 1
Course Arc & Why This Week Matters
Weeks 2–5 asked how organisms choose. Week 6 asks how organisms learn from experience in the first place—before they can form the values and predictions that drive decision-making. Habituation, sensitization, and perceptual learning are the brain's most fundamental mechanisms for filtering what deserves attention.
⟳ Course Arc Connection — Dopamine & Prediction Error
Before organisms can learn values (Weeks 7–8), they must decide what is worth attending to. Habituation is the brain's first attentional filter—deciding what to ignore. Sensitization signals "pay extra attention." These are prerequisites for associative and reinforcement learning. The comparator model of habituation is also a prediction-error model: respond when incoming stimuli do not match stored representations.
Section 2
Habituation
Habituation is a decrease in the strength or occurrence of a behavior after repeated exposure to the stimulus that produces it. It is the most widespread form of learning—observed in protozoa, sea slugs, and humans alike.
Classic paradigms: acoustic startle reflex in rats (repeated loud noise → declining jump amplitude); orienting response in infants (fixation time decreases with each re-presentation).
Six Key Properties — Know All of Them
| Property | What it means / exam implication |
|---|---|
| Stimulus Specificity | Habituation to stimulus A does not transfer to stimulus B, even in the same sensory modality. Rules out sensory fatigue. |
| Dishabituation | A novel or arousing stimulus restores the habituated response. Rules out motor exhaustion. |
| Spontaneous Recovery | Habituated response returns after a stimulus-free rest period. Habituation is not permanent. |
| Massed vs. Spaced | Massed → faster but less durable. Spaced → slower but longer-lasting. Parallels the spacing effect in explicit memory. |
| Below-Zero Habituation | Learning continues even after behavioral response reaches zero, revealed by delayed spontaneous recovery. This is latent learning. |
| Rate Depends on Arousal | Less arousing stimuli habituate more rapidly. Highly arousing stimuli may produce sensitization instead. |
⚠ Common Misconception
Habituation is NOT motor fatigue, sensory adaptation, or "getting bored." Evidence: stimulus specificity and dishabituation. If exhaustion were the cause, a novel stimulus should produce no orienting response—but it always does.
Section 3
Sensitization
Sensitization is the opposite of habituation: experiences with an arousing stimulus lead to stronger responses to subsequent stimuli. Unlike habituation, sensitization is NOT stimulus-specific—it broadly amplifies reactivity.
Classic example: Rats habituate to a repeated loud tone. If a subset then receives a foot shock, their startle response to the same tone jumps above baseline. The shock sensitized them.
Human measure: Skin conductance response (SCR / electrodermal activity). Underlies stress monitoring and classic lie detector tests.
Habituation vs. Sensitization: Key Contrast
| Feature | Habituation | Sensitization |
|---|---|---|
| Response direction | Decreases | Increases |
| Stimulus specificity | YES (specific) | NO (general) |
| Synaptic mechanism (Aplysia) | Homosynaptic depression (less glutamate) | Heterosynaptic facilitation (more glutamate via serotonin) |
| Arousal required? | No (works for low-arousal stimuli) | Yes (requires arousing stimulus) |
| Exposures needed? | Many repetitions | Sometimes a single intense event |
Section 4
Dual Process Theory
Proposed by Groves & Thompson (1970). Every stimulus presentation simultaneously activates two independent processes:
- Habituation process: Weakens the stimulus-response (S-R) connection through repeated activation.
- Sensitization process: Activates a "state system" that potentiates global reactivity.
- Observed behavior = NET result of both. Boring/non-arousing → habituation dominates. Highly arousing → sensitization dominates.
⟳ Connection to Prediction Errors
Dual process theory runs two parallel computations: how much to suppress the S-R connection AND how much to amplify arousal. This mirrors Rescorla-Wagner and TD frameworks (Weeks 7–8), where the brain simultaneously estimates expected value and computes prediction error.
Section 5
Neural Mechanisms: The Aplysia Model
Eric Kandel (Nobel Prize, 2000) used Aplysia californica (~20,000 neurons; individually identifiable) to find the first cellular mechanisms of habituation and sensitization.
Gill-Withdrawal Reflex Circuit
Siphon touch → Sensory neuron S (GLUTAMATE) → Motor neuron M → Gill muscles contract
| Feature | Habituation | Sensitization |
|---|---|---|
| Type | Homosynaptic | Heterosynaptic |
| Mechanism | Repeated activation of S → less glutamate released (synaptic depression) | Tail shock → interneuron IN → serotonin onto S terminals → more glutamate vesicles available |
| Scope | Only the S–M synapse (activated synapse) | Multiple synapses (S, U, others NOT activated by the shock) |
| Long-term structural change | Elimination (pruning) of S–M synapses | Formation of additional synaptic terminals (growth) |
📝 Exam Tip: Homo vs. Heterosynaptic
Homosynaptic = habituation = specific to the activated synapse = mirrors stimulus specificity.Heterosynaptic = sensitization = affects all synapses via serotonin interneuron = mirrors generality of sensitization.
Section 6
Perceptual Learning
Perceptual learning = repeated experience with stimuli makes them easier to distinguish. Unlike sensitization (amplifies responses), it specifically improves discriminability.
Real-world examples: chicken sexers (1 per half-second), dermatologists classifying rashes, wine tasters, musicians distinguishing instrument timbres.
| Route | Description | Key Feature |
|---|---|---|
| Mere Exposure Learning | Passive, incidental exposure improves discrimination without feedback. Gibson & Walk (1956) rat experiment. | Latent learning — revealed only when tested |
| Discrimination Training | Active training with feedback. Produces larger and faster perceptual changes and bigger cortical reorganization. | Largest cortical changes |
Three Theoretical Models
- Dual Process (Groves & Thompson): Shared features habituate more (seen twice as often); distinctive features remain salient.
- Comparator Model (Sokolov, Wagner): Brain stores representations and compares incoming stimuli. Orienting response on mismatch. Habituation = better representation = less mismatch.
- Differentiation Theory (E. Gibson): Representations start vague and gain detail with repeated exposure.
Note: Learning Specificity
A dog-show judge's expertise with cocker spaniels does NOT transfer to judging prize pigs. Training on vertical gratings does not transfer to horizontal ones.
Section 7
Priming & Familiarity
Priming: Prior exposure affects later responses without conscious awareness. Classic demo: word-stem completion ("CHI__" → "CHISEL" more likely if CHISEL was previously seen, even without conscious recall). Animal example: blue jays detecting camouflaged moths more accurately after recent exposure.
Familiarity: William James defined it as a "sense of sameness"—the perception of similarity when an event is repeated. Basis of recognition memory. Measured in rats and monkeys via the novel object recognition task.
Section 8
Spatial Learning & the Hippocampus
Tolman & Honzik (1930): Rats allowed to freely explore a maze (no food) for 10 days performed as well as reward-trained rats when food was introduced on day 11. They had formed a cognitive map. Key insight: spatial learning is latent—occurs without behavioral evidence until a test demands it.
Place Cells (John O'Keefe, Nobel Prize): Hippocampal neurons that fire only when a rat is at a specific location (their place field). Place cells use visual landmarks to define location.
- If maze AND landmark are rotated together → place fields rotate with them.
- If only the landmark is repositioned → place fields follow the landmark.
- Place fields shrink with repeated exploration — more precise spatial representation.
- London cab drivers (memorize 25,000 streets for "the Knowledge") have significantly larger hippocampi, correlating with years of experience.
⟳ Preview: Reinforcement Learning (Week 8)
Place cells encode STATES — the "s" in V(s) from value function learning. In model-based RL, knowing "what state am I in?" is crucial. The hippocampus answers this for spatial environments.
Section 9
Cortical Plasticity & Hebbian Learning
Receptive fields: Cortical neurons respond to specific stimulus features. In somatosensory cortex → body location (homunculus). In auditory cortex → sound frequency. Not proportional to body-part size—reflects use and sensitivity (thumbs and lips are over-represented).
- Repeated experience expands the cortical region dedicated to the trained stimulus.
- 2 hours of repeated fingertip stimulation → expanded somatosensory cortex representation AND improved two-point discrimination (fMRI + behavioral).
- Kittens with one eye sewn shut lose cortical representation for that eye permanently (developmental plasticity).
- Blind individuals show visual cortex activation during Braille reading—cross-modal plasticity.
Hebbian Learning — "Neurons that fire together, wire together"
"When an axon of cell A is near enough to excite cell B and repeatedly takes part in firing it, A's efficiency in firing B is increased." — Hebb (1949).Repeated coactivation → LTP → stable representation → pattern completion: a partial version of a familiar stimulus activates the full stored pattern. This explains priming.
Section 10
Clinical Applications
| Condition | Mechanism | Treatment / Relevance |
|---|---|---|
| Learned Non-Use (post-stroke) | Stroke damages sensory cortex → patient stops using affected limb (habituation to non-functional limb). Motor control is intact. | Constraint-Induced Movement Therapy (CIMT): bind the good arm → forced use = dishabituation + cortical reorganization in adjacent areas. |
| Cochlear Implants | Electrical stimulation produces "virtual" speech very different from natural sound. Users must learn to discriminate novel stimuli. | Perceptual learning curve: rapid at first, gradual over years. Most effective in young children and recently deafened adults. Mechanism: cortical plasticity in auditory cortex. |
Key Terms — 25 Flashcards
Click any card to reveal its definition. Use the filters to focus on a category.
Behavioral
Habituation
Click to reveal →
Behavioral
A decrease in the strength or occurrence of a behavior after repeated exposure to the stimulus that produces it. Stimulus-specific and reversible.
Behavioral
Sensitization
Click to reveal →
Behavioral
Experiences with an arousing stimulus lead to stronger responses to subsequent stimuli. General—NOT stimulus-specific.
Behavioral
Dishabituation
Click to reveal →
Behavioral
Recovery of a habituated response following presentation of a novel or arousing stimulus. Rules out motor exhaustion.
Behavioral
Spontaneous Recovery
Click to reveal →
Behavioral
Reappearance of a habituated response after a period without stimulus presentation. Shows habituation is not permanent.
Behavioral
Latent Learning
Click to reveal →
Behavioral
Learning that occurs without immediate behavioral evidence; revealed only by an appropriate test. See: Tolman's maze rats, below-zero habituation, mere exposure learning.
Behavioral
Perceptual Learning
Click to reveal →
Behavioral
Repeated experience with a set of stimuli improves the organism's ability to distinguish those stimuli. Can occur passively.
Behavioral
Mere Exposure Learning
Click to reveal →
Behavioral
Perceptual learning through passive exposure to stimuli, without explicit feedback. A form of latent learning.
Behavioral
Discrimination Training
Click to reveal →
Behavioral
Perceptual learning with feedback about accuracy; associated with the largest cortical reorganization.
Behavioral
Priming
Click to reveal →
Behavioral
Prior exposure to a stimulus (even without conscious awareness) facilitates later recognition or processing of that stimulus.
Behavioral
Familiarity
Click to reveal →
Behavioral
William James's "sense of sameness"—the perception of similarity when an event is repeated. Basis of recognition memory. Measured via novel object recognition task.
Neural
Homosynaptic Depression
Click to reveal →
Neural
Reduced neurotransmitter (glutamate) release with repeated activation of the same synapse; the cellular mechanism of short-term habituation in Aplysia.
Neural
Heterosynaptic Facilitation
Click to reveal →
Neural
Enhancement of multiple synapses (including unactivated ones) via interneuron serotonin release. Mechanism of Aplysia sensitization.
Neural
Cortical Plasticity
Click to reveal →
Neural
Experience-dependent changes in the receptive fields and topographic maps of sensory cortices. More training → more cortical real estate.
Neural
Hebbian Learning
Click to reveal →
Neural
"Neurons that fire together, wire together." Coactivity strengthens synaptic connections via LTP. Basis of pattern completion and priming.
Neural
Receptive Field
Click to reveal →
Neural
The range of stimuli that causes a particular cortical neuron to fire. Narrows with repeated exposure (perceptual learning).
Neural
Topographic Map
Click to reveal →
Neural
Organization of cortex in which neighboring neurons respond to similar stimuli (e.g., somatosensory homunculus). Reflects use, not anatomy.
Theoretical
Dual Process Theory
Click to reveal →
Theoretical
Groves & Thompson (1970). Every stimulus activates both a habituation process (weakens S-R) and a sensitization process (potentiates global reactivity). Observed behavior = net result.
Theoretical
Comparator Model
Click to reveal →
Theoretical
Sokolov / Wagner. Brain compares incoming stimuli to stored representations; an orienting response is triggered on mismatch. Habituation = better representation = less mismatch.
Theoretical
Differentiation Theory
Click to reveal →
Theoretical
E. Gibson. Representations start vague and become more detailed with repeated exposure as more features are encoded. Explains why first exposure captures general form only.
Spatial
Place Cells
Click to reveal →
Spatial
Hippocampal neurons that fire maximally when an animal is at a specific location (place field). Place fields shrink with experience, become more precise.
Spatial
Cognitive Map
Click to reveal →
Spatial
Tolman's term for an internal representation of spatial layout formed through exploration. Revealed when a reward is introduced (latent learning).
Spatial
Novel Object Recognition
Click to reveal →
Spatial
Task in which preference for exploring a novel over a familiar object indexes recognition memory. Hippocampal lesions impair performance when context or position must be integrated.
Clinical
Learned Non-Use
Click to reveal →
Clinical
Habituation to a desensitized limb following stroke; patient stops using it despite intact motor control. Treated with Constraint-Induced Movement Therapy (CIMT).
Clinical
Cochlear Implant
Click to reveal →
Clinical
Sensory prosthesis that electrically stimulates auditory nerves. Users must acquire new perceptual learning to interpret "virtual" speech sounds via cortical plasticity.
Neural
Synaptic Depression
Click to reveal →
Neural
Reduced glutamate vesicle release with repeated activation. The cellular mechanism of short-term habituation in Aplysia. Distinct from the synapse pruning of long-term habituation.
Practice Multiple Choice — 20 Questions
Click an option to check your answer. Explanations appear automatically.
34
Questions
0
Correct
0
Incorrect
—
Score
Big Picture Synthesis
How the week's concepts connect across levels of analysis and to the course arc.
The Unifying Principle
All phenomena this week share a common architecture: the brain tracks regularities in the environment and adjusts its responses accordingly. Adjustments can go down (habituation), up (sensitization), or toward finer discrimination (perceptual learning). Mechanisms span from single synapses to whole-brain reorganization.
Levels of Analysis
Behavior
Response amplitude decreases (habituation)
Response amplitude increases (sensitization)
Discrimination improves (perceptual learning)
Circuit
Aplysia gill-withdrawal circuit (S → M)
Hippocampal place cells
Cortical topographic maps
Synapse
Glutamate vesicle depletion (habituation)
Serotonin modulation (sensitization)
LTP / Hebbian (perceptual learning)
Structure
Synapse pruning (long-term habituation)
Synapse growth (long-term sensitization)
Cortical map expansion (perceptual learning)
Likely Exam Themes
- Distinguish habituation, sensitization, dishabituation, and spontaneous recovery from scenario descriptions
- Apply homosynaptic vs. heterosynaptic to Aplysia circuits and link to behavioral specificity
- Identify the mechanism behind an experimental result (synaptic depression? cortical plasticity? Hebbian?)
- Distinguish mere exposure vs. discrimination training and which produces larger cortical changes
- Interpret place cell data—what happens when landmarks move
- Explain clinical phenomena (learned non-use, cochlear implants) using the week's concepts
- Connect this week's content back to prediction error and the RL arc
Cross-Course Connections
- Comparator model = prediction error: respond to mismatch between stimulus and stored representation
- Dual process theory mirrors RL: parallel computation of S-R suppression and arousal amplification
- Place cells = state representation for V(s) in model-based RL (Week 8)
- Latent learning = policy-free value learning: knowledge acquired without behavioral expression
- Hebbian LTP is the synaptic substrate for many forms of associative learning (Weeks 7–8)
- Perceptual learning specificity parallels the stimulus generalization problem in RL