Two-Week Module — Know Both Halves
Week 12 (Chapter 10) covers emotion theories, CERs, learned helplessness, emotional memory, and amygdala circuits. Week 13 (Chapter 11) covers social learning mechanisms, mirror neurons, birdsong, and ASD. Both chapters appear on the exam and connect through shared circuits (dopamine, basal ganglia, PFC).
Section 1
Course Arc: Where Module 12 Fits
Emotion (amygdala) = the urgency signal that decides which events get cemented into long-term memory. Without emotional tagging, most experiences are forgotten within days.
Social learning = the highest-bandwidth information channel available to an intelligent agent: rather than accumulating errors across thousands of personal experiences, you inherit others' compressed outcomes in minutes.
Both chapters revisit familiar circuits — basal ganglia, hippocampus, PFC — in new emotional and social roles. The key conceptual move: these are not new systems, they are old systems operating under new modulatory conditions.
Connect Backward
Dopamine updates predictions (Weeks 5–8); amygdala gates which predictions are worth updating by ensuring survival-relevant events are remembered after one trial. Social learning is model-based RL where the model includes representations of other agents' value functions — you inherit their Q-values instead of learning from scratch.
- Week 12 → Chapter 10: Emotion theories, conditioned fear, learned helplessness, amygdala substrates.
- Week 13 → Chapter 11: Social learning taxonomy, mirror neurons, birdsong circuits, clinical applications.
- Film bridge: Inside Out — emotions as characters competing to control behavior (three-component model made visible); The Social Network — social learning as the engine of cultural accumulation.
Section 2
What Is Emotion? Three-Component Model
| Component | Description | Examples |
|---|---|---|
| Physiological Response | ANS-mediated; automatic arousal of body systems | Heart rate increase, perspiration, epinephrine release, cortisol, piloerection |
| Overt Behavior | Externally observable motor output of emotional state | Freezing, fleeing, facial expressions, approach, piloerection (fur/hair standing) |
| Conscious Feeling | Subjective experience; depends on PFC and cognitive appraisal | Feeling afraid, feeling joyful — the "what it is like" component |
Universal emotions (Ekman): Happiness, Sadness, Anger, Fear, Disgust, Surprise — recognized cross-culturally with consistent facial expressions. Display rules (social norms about when to show emotion) vary across cultures, but the underlying physiological and behavioral states are universal.
Fight-or-flight vs. tend-and-befriend: Classic stress response mobilizes for conflict or escape; tend-and-befriend (more common in females) mobilizes social affiliation and caregiving as stress coping.
Critical fact: Epinephrine (adrenaline) cannot cross the blood-brain barrier. The brain analogue is norepinephrine, which is released centrally from the locus coeruleus and plays the key role in emotional memory modulation.
Lecture Demo — Know This
Suppressed smile leading to laughter: demonstrates the James-Lange direction (body state feeds back to produce/amplify feeling). PFC acts as a top-down brake on emotional expression — damage releases inhibition, flooding behavior.
Section 3
Theories of Emotion
| Theory | Causal Sequence | Key Evidence |
|---|---|---|
| James-Lange | Stimulus → bodily response → conscious feeling follows (body causes feeling) | Facial feedback hypothesis: manipulating facial expression changes reported emotion; suppressed smile → laughter in lecture demo |
| Cannon-Bard | Emotional stimulus → SIMULTANEOUS bodily response AND conscious feeling (parallel, not sequential) | Criticizes James-Lange: visceral responses are slow and similar across emotions; proposed thalamus routes signal to both body and cortex at once |
| Two-Factor (Schachter & Singer) | Physiological arousal + cognitive appraisal → emotion label and experience | Capilano Bridge study: physical arousal from swaying bridge misattributed as attraction to interviewer; epinephrine injection + social context manipulates reported emotion |
Two-Factor Theory — The Key Insight
Arousal is undifferentiated (it's just activation). The LABEL applied to that arousal depends on the cognitive context available. Same arousal + different context = different emotion. This is why physical exertion can feel like attraction (misattribution of arousal) and why context is so important in emotional experience.
Section 4
Conditioned Emotional Responses, Escape & Avoidance
Conditioned Emotional Responses (CERs): A single CS-US pairing can produce robust fear. CERs are highly resistant to extinction and readily generalize to related stimuli. This reflects the adaptive value of rapid, broad, sticky fear learning.
| Type | Mechanism | Why It Persists |
|---|---|---|
| Conditioned Escape | Operant — response terminates an ongoing aversive stimulus (negative reinforcement) | US actually present; each escape confirms relief |
| Conditioned Avoidance | Mowrer two-factor: (1) classical: CS acquires fear; (2) operant: response to CS reduces CS-fear (US prevented) | US never arrives → CS-US association never extinguished → fear persists → avoidance persists indefinitely |
The Avoidance Paradox — High-Yield Exam Topic
The lever press/jump is reinforced by reduction of conditioned fear to the CS — not by US termination. Since the US never occurs, the CS-US association is never updated by extinction. The organism can never "learn" the US is gone because it never waits long enough for that information. This is why conditioned avoidance is essentially self-perpetuating and why PTSD avoidance behaviors are so intractable.
Clinical Bridge: PTSD
PTSD avoidance = conditioned avoidance. Trauma-related CS-like stimuli are avoided; the US (trauma) never re-occurs; extinction is therefore impossible as long as avoidance is maintained. Treatment (exposure therapy) must break the avoidance cycle to allow extinction of the CS-fear.
Lecture Demo: Fear Conditioning Time Course
Tone (CS) + shock (US) → blood pressure spikes within secondsConditioned response (BP to tone alone) observable after 2–3 pairings
= single session, highly rapid acquisition
Section 5
Learned Helplessness
Seligman & Maier (1967): Dogs pre-exposed to inescapable shock lay down in the shuttle box later and did not try to escape — even when escape was easy. They had learned that responses are ineffectual.
Definition: Learned helplessness = an expectancy that responses are ineffectual, leading to reduced motivation to attempt new avoidance or escape behaviors in novel contexts. It is a cognitive/motivational phenomenon, not a motor failure.
- Immunization: Prior experience with escapable shock confers resilience — animals pre-exposed to controllable aversives do not become helpless after subsequent inescapable shock.
- Depression link: Antidepressants that work in humans also reverse helplessness in rats. Anhedonia and motivational withdrawal in helpless animals mirror clinical depression phenotype.
- Key insight: Physically carrying the helpless dog over the barrier did not help — the animal had to be carried repeatedly before it re-learned that its own actions could succeed. The expectancy must be updated by controllability experience.
Lecture: Morris Water Maze Analog
When the hidden platform is removed, mice stop searching in the target quadrant — a form of acquired futility. Changing the environment (novel room, new cues) re-engages motivation by resetting the expectancy in a new context. This shows the context-specificity of helplessness and its reversibility.
Section 6
Emotion and Memory
Cahill et al. (1995): Emotional arousal selectively enhances encoding of emotionally significant material but NOT surrounding neutral content. Epinephrine injection after training improves memory (modulation window closes at ~120 min). Propranolol (norepinephrine blocker) eliminates the memory enhancement for emotional material.
Mood congruency: Current emotional state functions as a retrieval cue — positive mood retrieves positive memories, negative mood retrieves negative memories. This can create self-reinforcing cycles (depression).
| Phenomenon | Finding | Mechanism |
|---|---|---|
| Flashbulb memories | Vivid, high-confidence memories for emotional events; but NOT necessarily accurate (Talarico & Rubin 2003 — 9/11 memories degraded while confidence stayed high) | Amygdala tags memory as important, producing subjective sense of permanence; but hippocampal content still undergoes normal decay and reconsolidation errors |
| Reconsolidation | Reactivating a memory opens a labile window where the trace can be modified before restabilizing | Therapeutic implication: propranolol during reconsolidation window can reduce emotional charge without erasing factual content |
| Cortisol inverted-U | Low-moderate cortisol facilitates LTP and memory consolidation; high/chronic cortisol impairs hippocampal function | Prolonged stress causes dendritic atrophy in hippocampus; consistent with hippocampal volume reduction in chronic PTSD |
Propranolol Therapeutic Window
Propranolol blocks beta-adrenergic receptors (norepinephrine). Given before an emotional story, it eliminates the memory advantage for emotional portions (Cahill et al. 1994). Given at reconsolidation (when a PTSD memory is reactivated), it can dampen the BLA-mediated emotional component while leaving the factual hippocampal record intact. This is the basis of a proposed PTSD treatment.
Section 7
Brain Substrates of Emotional Learning
| Structure | Role | Key Evidence |
|---|---|---|
| Lateral nucleus (amygdala) | Primary input gate; site of Hebbian LTP for CS-US association formation | LTP required at lateral nucleus synapses during CS-US pairing; lateral nucleus lesions prevent fear acquisition |
| Central nucleus (amygdala) | Output hub — drives ANS responses and motor fear expression (freezing, startle, BP changes) | Central nucleus lesions abolish conditioned freezing and SCR despite intact lateral nucleus |
| Basolateral amygdala (BLA) | Modulates memory consolidation in hippocampus and cortex via norepinephrine projections | BLA lesions eliminate epinephrine's memory-enhancing effect; propranolol in BLA mimics BLA lesion |
| Hippocampus | Encodes contextual information (WHERE fear was learned); contextual fear conditioning requires hippocampus | Hippocampal lesions abolish contextual fear but spare cued (tone) fear |
| Medial PFC (mPFC) | Inhibits amygdala top-down; required for extinction; damage → disinhibited emotional expression OR blunting | Extinction requires intact mPFC; vmPFC activity inversely correlates with amygdala during extinction recall |
LeDoux's Two Pathways to Amygdala
FAST path: Thalamus → Lateral amygdala (~12 ms) — coarse, rough, rapid response before full identificationSLOW path: Thalamus → Sensory cortex → Lateral amygdala (~19 ms) — detailed, allows cancellation if stimulus is safe
Both paths converge on lateral nucleus → CS-US association stored there
Bechara Double Dissociation (1995) — High-Yield
Amygdala damage: impairs conditioned SCR (skin conductance response) but preserves declarative knowledge ("CS was paired with shock"). Hippocampal damage: impairs declarative knowledge but preserves conditioned SCR. Emotional learning and episodic memory of emotional events use different systems. This is the strongest direct evidence for the dual-system model.
Section 8
Clinical: Phobias and PTSD
| Phobia | PTSD | |
|---|---|---|
| Core mechanism | Excessive CER that has generalized to broad range of CS-like stimuli | Failure of extinction after trauma; CER remains strong months/years later |
| Acquisition | Can be directly conditioned OR acquired observationally (watching another's fear response) | Single traumatic CS-US pairing; CRs acquired faster, extinguished more slowly in PTSD patients |
| Predisposing factor | Prior sensitization, genetic vulnerability | Smaller hippocampal volume (Gilbertson 2002 twin study — non-deployed co-twins also had smaller hippocampi → predisposition, not consequence) |
| Treatment | Systematic desensitization — graded non-reinforced CS exposure | Exposure therapy; propranolol at reconsolidation; VR therapy |
Fear Extinction ≠ Erasure
Fear extinction is new learning (inhibitory CS-noUS association), not erasure of the original CS-US memory. The original memory persists and can be reinstated from the original training context. This is why PTSD patients relapse when returning to the trauma environment — the extinction memory is context-specific but the original fear memory is not.
Gilbertson (2002) Twin Study — Causal Interpretation
Vietnam veterans WITH PTSD and their non-deployed identical twins BOTH had smaller hippocampal volumes. Since the non-deployed twins never experienced combat, smaller hippocampal volume must be a predisposing vulnerability, not a trauma consequence. This reframes PTSD risk as partly a biological predisposition.
Section 9
Social Learning: Behavioral Mechanisms
Social learning: Learning by actively monitoring events involving other individuals. No direct reinforcement of the learner is required. This is the key distinction from standard operant conditioning.
| Type | Definition | Example / Evidence |
|---|---|---|
| True imitation | Copies the SPECIFIC MOTOR ACT of the model (not just the outcome) | Children in Whiten (1996) artificial fruit task copied poking motion; counterintuitively, children outperform chimps at true imitation |
| Emulation | Copies the GOAL/OUTCOME; method is flexible (observer uses own means) | Chimps used palms rather than index finger despite watching the model poke — they achieved the outcome without copying the movement |
| Contagion | Innate, reflexive, automatic response to same response in conspecifics; NOT social learning; no episodic memory storage | Yawning, laughing — no learning occurs; cannot be suppressed voluntarily in the way imitation can |
| Observational conditioning | Classical conditioning applied to OBSERVED events — acquires emotional/fear response by watching another organism's CR or UR | Mineka & Cook (1988): lab-reared monkeys learned snake fear by watching wild conspecifics react; single observation sufficient |
| Stimulus enhancement | Another organism's action directs the observer's attention to a stimulus; observer may then learn about that stimulus on their own | Watching someone use a tool directs attention to the tool; observer independently learns its properties — NOT motor copying |
Bandura's Four Conditions — Know All Four
1. Attention — observer must attend to the model. 2. Retention — observed behavior must be encoded in memory. 3. Reproduction capacity — observer must be physically able to perform the action. 4. Motivation — some reason to perform (can be vicarious reinforcement). Direct reinforcement of the learner is NOT required — this is Bandura's key departure from Skinnerian behaviorism.
Template Model of Birdsong (Three Phases)
Phase 1: MEMORIZE — juvenile listens to adult tutor, encodes auditory templatePhase 2: SENSORIMOTOR PRACTICE — sings alone, compares output to template; TD learning drives convergence
Phase 3: SOCIAL USE — performs for female; dopamine prediction error gated OFF during performance (female's response = new teaching signal)
Goldberg lab (zebra finch): Dopamine performance prediction error in Area X present during solo singing (Phase 2 rehearsal) but disappears when a female is present. Context gates two modes: rehearsal uses internal template comparison; performance uses social feedback as the learning signal. These modes are mutually exclusive and context-dependent.
Section 10
Neural Substrates & Clinical Perspectives in Social Learning
Mirror neurons (Rizzolatti, macaque area F5): Fire both when the monkey performs an action AND when it observes the same action performed by another. Some neurons are action-selective (fire for one specific movement); others are goal-selective (fire for any action achieving a particular outcome).
Direct-matching hypothesis: Stored motor action representations automatically map observed actions onto the observer's own motor system — providing a neural substrate for understanding others' actions by simulating them internally.
| Region / Finding | Function |
|---|---|
| Area F5 / inferior frontal gyrus | Primary mirror neuron locus in macaques; human homologue (IFG) shows overlapping activation for action execution and observation |
| Goal-selective mirror neurons | Fire for ANY action achieving the outcome — provides neural basis for emulation (copying goal, not specific act) |
| Human evidence | Mu rhythm suppression (EEG during observation), TMS motor-evoked potentials elevated during action observation, fMRI IFG overlap |
| Frontal lobe (voluntary control) | Controls WHEN to engage copying; frontal damage → increased automatic imitation + impaired voluntary imitation |
Birdsong Circuits — Mammalian Analogs
HVC (timing) ≅ motor cortexRA (fine features; mirror-neuron-like) ≅ primary motor output
Area X (dopamine learning, basal ganglia analogue) ≅ dorsal striatum
LMAN (adds variability/exploration) ≅ prefrontal/frontal cortex
Rat food preference (Galef & Wigmore, 1983): A single 15-min social interaction (demonstrator rat with food on breath) produced months-long food preference in observer rats. Hippocampal lesions disrupted this with a retrograde gradient (lesions immediately after interaction = worst impairment) — confirming hippocampal-dependent episodic encoding of social information.
ASD and Frontal Stroke — The Unifying Insight
Both ASD and frontal stroke show increased automatic imitation (echolalia in ASD; automatic mirroring in frontal patients) combined with impaired voluntary imitation. This paradox is explained by the dual-circuit model: automatic imitation = mirror neuron activation (intact); voluntary imitation = mirror neurons + frontal gate deciding WHEN to copy (impaired). The "broken mirrors" hypothesis for ASD is now considered overstated — mirror neurons may be intact but the frontal control circuit is disrupted.
Key Terms — 25 Flashcards
Click any card to reveal its definition. Use the filters to focus on a category.
Theory
Three-component model of emotion
Click to reveal →
Theory
Emotion = cluster of physiological responses (heart rate, hormones) + overt behaviors (freezing, facial expressions) + conscious feelings (subjective experience). All three components occur together.
Theory
James-Lange theory
Click to reveal →
Theory
Emotional stimulus → bodily response first → conscious feeling follows from perception of that body state. "We feel afraid because we tremble." Evidence: facial feedback — forcing a smile produces reported positive emotion.
Theory
Cannon-Bard theory
Click to reveal →
Theory
Emotional stimulus evokes simultaneous physiological response AND conscious feeling — parallel, not sequential. Thalamus routes signal to both body and cortex at the same time. Critiques James-Lange's sequential model.
Theory
Two-factor theory (Schachter & Singer)
Click to reveal →
Theory
Emotion = physiological arousal + cognitive appraisal. Arousal is undifferentiated; the cognitive context supplies the label. Capilano Bridge: bridge-induced arousal misattributed as attraction when interviewer is present.
Theory
Direct-matching hypothesis
Click to reveal →
Theory
Motor action memories are stored in circuits that simultaneously encode visual and motor representations. Observing an action automatically activates the same motor representation used to perform it — the neural basis of imitation and action understanding.
Behavioral
Conditioned emotional response (CER)
Click to reveal →
Behavioral
Classically conditioned fear response to a CS. Key properties: fast acquisition (single pairing possible), broad generalization, highly extinction-resistant. Reflects the adaptive value of rapid fear learning for survival.
Behavioral
Conditioned escape
Click to reveal →
Behavioral
Operant response that terminates an ongoing aversive stimulus. Reinforced by negative reinforcement (relief from present US). US actually occurs; each escape confirms effectiveness of the response.
Behavioral
Conditioned avoidance
Click to reveal →
Behavioral
Operant response to a warning CS that prevents US occurrence. Extremely persistent: reinforced by CS-fear reduction; since US never comes, CS-fear never extinguishes; therefore avoidance never extinguishes. Mowrer two-factor mechanism.
Behavioral
Learned helplessness
Click to reveal →
Behavioral
Expectancy that responses are ineffectual, leading to reduced motivation for new avoidance/escape attempts. Induced by inescapable aversives; reversed by controllability experience; parallels depression's anhedonia and motivational withdrawal.
Behavioral
Flashbulb memory
Click to reveal →
Behavioral
Vivid episodic memory for an emotionally significant event; accompanied by high confidence. But NOT necessarily accurate — Talarico & Rubin (2003) showed 9/11 memory details degraded over months while confidence remained high. Vividness ≠ accuracy.
Behavioral
Mood congruency
Click to reveal →
Behavioral
Current emotional state biases memory retrieval toward memories of the same valence. Positive mood retrieves positive memories; negative mood retrieves negative memories. Can create self-reinforcing cycles in depression.
Behavioral
Social learning
Click to reveal →
Behavioral
Learning by monitoring other individuals' actions and outcomes. No direct reinforcement of the learner required. Allows inheriting compressed outcomes of others' experiences — the highest-bandwidth information channel for intelligent agents.
Behavioral
True imitation
Click to reveal →
Behavioral
Copying the specific motor acts of a model, not just the goal. Demonstrated by children in Whiten's artificial fruit task (copied poking motion). Counterintuitively, human children are better true imitators than chimps.
Behavioral
Emulation
Click to reveal →
Behavioral
Copying the goal/outcome without replicating the specific motor acts. Chimps in Whiten task used palms rather than index fingers — achieving the result their own way. Goal-selective mirror neurons provide the neural basis.
Behavioral
Contagion
Click to reveal →
Behavioral
Innate, reflexive, automatic response to observing the same response in conspecifics (yawning, laughing). NOT social learning — no episodic memory storage, no intentional copying. Cannot be voluntarily suppressed in the way imitation can.
Behavioral
Observational conditioning
Click to reveal →
Behavioral
Acquiring an emotional/fear response by watching another organism's CR or UR. Classical conditioning applied to observed events. Mineka & Cook (1988): lab monkeys acquired snake fear after watching wild conspecifics react. Single observation sufficient.
Neural
Reconsolidation
Click to reveal →
Neural
Reactivating a stored memory renders it transiently labile — it can be modified before restabilizing. The reconsolidation window is the therapeutic target: propranolol given during reactivation can reduce the emotional charge (BLA-mediated) while leaving factual content (hippocampal) intact.
Neural
Lateral nucleus (amygdala)
Click to reveal →
Neural
Primary input gate of the amygdala; site of Hebbian LTP for CS-US association formation. Receives both the fast thalamic path and the slow cortical path. Lesions prevent fear acquisition without affecting fear expression driven by the central nucleus.
Neural
Central nucleus (amygdala)
Click to reveal →
Neural
Output hub of the amygdala: drives ANS (heart rate, blood pressure) and motor fear expression (freezing, startle, piloerection). Lesions abolish conditioned SCR and freezing even when the lateral nucleus CS-US association is intact.
Neural
Basolateral amygdala (BLA)
Click to reveal →
Neural
Modulates memory consolidation in hippocampus and cortex via norepinephrine projections. BLA activation during emotional events is the mechanism by which arousal selectively enhances memory for emotionally significant content. Propranolol in BLA eliminates this enhancement.
Neural
LeDoux two pathways
Click to reveal →
Neural
Fast (~12ms): thalamus → lateral amygdala directly — coarse but rapid; enables response before cortical identification is complete. Slow (~19ms): thalamus → sensory cortex → lateral amygdala — detailed; allows cancellation if stimulus is safe. Speed-accuracy tradeoff at the neural level.
Neural
Mirror neurons
Click to reveal →
Neural
Neurons that fire both when the animal performs an action AND when it observes the same action in another. Found in macaque premotor area F5 (Rizzolatti). Goal-selective mirror neurons fire for any action achieving an outcome — basis for emulation. Human evidence: mu rhythm suppression, TMS MEPs, fMRI IFG overlap.
Neural
Template model (birdsong)
Click to reveal →
Neural
Phase 1: Juvenile memorizes auditory template from tutor. Phase 2: Sensorimotor practice — sings alone, compares output to template; TD learning (dopamine in Area X) drives convergence. Phase 3: Social use — performs for female; dopamine prediction error gated off; female's response is the new teaching signal.
Clinical
Systematic desensitization
Click to reveal →
Clinical
Phobia treatment: graded, non-reinforced CS exposure to extinguish the CER. Patient is exposed to progressively fear-inducing stimuli while US never occurs, allowing extinction. Extinction = new learning (CS-noUS), not erasure; original CS-US memory persists but is inhibited by extinction memory.
Clinical
PTSD
Click to reveal →
Clinical
Failure of extinction after trauma. PTSD patients acquire fear CRs faster and extinguish more slowly. Symptoms: hyperarousal, flashbacks, avoidance persist months/years. Predisposing factor: smaller hippocampal volume (Gilbertson 2002 twin study — vulnerability, not consequence). Treatment: exposure therapy, propranolol at reconsolidation.
Practice Multiple Choice — 20 Questions
Click an option to check your answer. Explanations appear automatically.
60
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
The amygdala is the salience detector that decides which events dopamine should prioritize for encoding. Social learning is model-based RL where the model includes other agents' value functions — you inherit their compressed Q-values in minutes instead of learning from trial and error across years. Both mechanisms exploit the same underlying circuits (BG, hippocampus, PFC) but under new modulatory conditions.
Levels of Analysis
Behavior
3-component model: physio + behavioral + conscious
CERs: fast, broad, sticky fear learning
Social learning ≠ conditioning; no direct RL needed
Circuit
Amygdala: lateral (learn), central (express), BLA (modulate)
Hippocampus: context encoding, one-shot social memory
Mirror neurons + frontal gate = social copying control
Synapse
LTP at lateral amygdala for CS-US association
NE from BLA → hippocampal LTP enhancement
Dopamine in Area X: TD error during birdsong rehearsal
Clinical
PTSD = extinction failure; avoidance perpetuates
Propranolol at reconsolidation window = PTSD treatment
ASD/frontal stroke: auto imitation up, voluntary down
Likely Exam Themes
- Three-component model vs. universal emotions (Ekman)
- James-Lange vs. Two-factor theory — Capilano Bridge
- Avoidance paradox — why it persists despite no US; PTSD link
- Learned helplessness — expectancy mechanism; immunization; depression
- Bechara double dissociation — amygdala SCR vs. hippocampus declarative
- LeDoux dual pathway — fast vs. slow to lateral amygdala
- Gilbertson twin study — hippocampal volume as predisposition
- True imitation vs. emulation — Whiten task; chimps vs. children
- Bandura's four conditions — no direct reinforcement required
- Mirror neurons + frontal gate — ASD and frontal stroke paradox
Cross-Course Connections
- Amygdala = salience detector determining what dopamine should prioritize (Weeks 5–8)
- BLA → NE → hippocampal LTP = same LTP mechanism from Week 3 applied to emotional encoding
- Avoidance paradox = extinction failure from Week 6 applied to clinical context
- Birdsong Area X ≅ basal ganglia TD learning from Week 8 — same architecture
- Dopamine gated off during social performance = context-switching in prediction error signals
- Social learning = model-based RL (Week 11 habits vs. goals) with other agents as the model
- Same dopamine prediction error from Weeks 5–8 now shown as the WM stability modulator (Week 11) and birdsong learning signal (Week 13)