Lecture 10(AIver) - Emotion

Patient SM's condition

Patient SM has Urbach-Wiethe disease — a rare recessive genetic disorder. About 50–75% of cases show bilateral symmetrical calcifications on the medial temporal lobes, destroying the amygdala.

What emotion could Patient SM NOT recognise?

Fear. She could accurately identify anger, disgust, happiness, sadness, and surprise, but could not recognise fear from facial expressions, despite knowing what fear was conceptually.

What does Patient SM's case demonstrate about the amygdala?

Amygdala damage leads to impairment in fear recognition specifically — the landmark conclusion of Adolphs et al. (1994).

Feinstein et al. (2011) — Patient SM fear tasks

Patient SM showed a complete lack of fear across three tasks: (1) holding a snake at an exotic pet store for 3+ min, (2) visiting Waverly Hills Sanatorium (haunted house) at Halloween, (3) watching clips from horror films (Blair Witch, The Ring, The Shining, Halloween).

Izard (2010) definition of emotion

Emotion consists of neural circuits (partially dedicated), response systems, and a feeling state/process that motivates and organises cognition and action. It provides information, may include cognitive appraisals, expressions/social-communicative signals, and may motivate approach or avoidant behaviour.

Three components all emotion researchers agree on

(1) Physiological reaction to a stimulus, (2) Behavioural response, (3) Feeling/subjective experience.

Affect vs Emotions vs Moods — key distinctions

Affect = broad range of feelings (includes emotions and moods). Emotions: caused by specific events, very brief (seconds/minutes), specific (e.g. fear, anger), distinct facial expressions, action-oriented. Moods: cause often unclear, last longer (hours/days), more general (positive/negative affect), cognitive in nature.

Conventional view of emotion generation

Stimuli → cognitive, motivational, somatic, and behavioural responses → emotional experience + subjective feelings → verbal report.

Anderson & Adolphs (2014) model of emotion

Stimuli → central emotion state → multiple outputs: observed behaviour, subjective reports, psychophysiology, cognitive changes, somatic responses (with feedback loops).

Locationist vs Psychological Constructionist accounts of emotion

Locationist: each discrete emotion (e.g. fear, anger) is biologically basic, inherited, and localised to a specific brain region/network. Constructionist: emotions emerge from more basic psychological operations not specific to emotion; some brain regions are commonly activated by different emotions.

Analogy for Locationist vs Constructionist accounts

Locationist: anger = salt, happiness = sugar (single pure ingredient). Constructionist: anger = curry, happiness = cake (complex mix of shared basic ingredients).

Ekman's six basic emotions

Anger, Fear, Disgust, Surprise, Happiness, Sadness.

Ekman's criteria for basic emotions (TABLE 10.2)

Distinctive universal signals, presence in other primates, distinctive physiology, distinctive universals in antecedent events, rapid onset, brief duration, automatic appraisal, unbidden occurrence. (1999 additions: distinctive appearance developmentally, distinctive thoughts/memories/images, distinctive subjective experience.)

Are basic emotions universal? — Matsumoto & Willingham (2006) Olympic study

84 judo athletes from 35 countries at 2004 Athens Olympics. At match completion (2.5–3s): winners showed Duchenne smiles (92% gold, 70% bronze); losers showed sadness, contempt, or nothing. Culture did NOT impact this immediate response. On the podium: large cultural differences emerged (North American/East Asian winners smiled; other countries showed more variability).

Complex emotions — definition and example

Complex emotions are combinations of basic emotions that may be socially or culturally learned. Example: Jealousy = emotional response to a real or perceived relationship threat (involves increased left frontal cortical activity).

Dimensional theories of emotion — two dimensions

(1) Valence and Arousal: emotions mapped on a 2D space (negative–positive valence × low–high arousal). (2) Approach vs Withdrawal: left hemisphere = approach motivation; right hemisphere = avoidance/withdrawal motivation.

James-Lange Theory of emotion

Stimulus → conscious perception → physiological response (adrenaline, increased HR, sweating) → behavioural response → nonconscious cognitive interpretation of physiology → subjective emotional feeling. Emotion follows from bodily changes.

Cannon-Bard Theory / LeDoux's Fast and Slow Roads

LeDoux builds on Cannon-Bard (parallel processing). Seeing a stimulus splits into: SLOW road — cortical cognition ("that looks like a dangerous bear") → emotional feeling; FAST road — hardwired subcortical defensive circuit → fight-or-flight behaviour. Both occur in parallel.

Appraisal Theory (Lazarus)

Cognition is central. Stimulus → quick risk-benefit cognitive appraisal → "I am in danger" → emotional feeling → behavioural response. Cognition precedes emotion.

Panksepp's view of emotion generation

Stimulus → subcortical processing → core emotion → (a) instinctive emotional behaviour supported physiologically (e.g., running) AND (b) emotional feeling ("I'm scared").

Limbic system — who described it and what structures are included?

James Papez proposed the circuit theory (hypothalamus, anterior thalamus, cingulate gyrus, hippocampus). Paul MacLean expanded it to the limbic system, adding medial cortex, basal ganglia, amygdala, and orbitofrontal cortex. "Limbic" = Latin for rim (structures form a rim around the corpus callosum). This work spawned locationist views.

The amygdala — location and structure

Small almond-shaped structures in the medial temporal lobe adjacent to the anterior hippocampus. Contains 12 nuclei grouped into 3 complexes: (1) Basolateral (B, La) — receives/translates sensory info into instrumental behaviours; (2) Centromedial (Ce) — connected to brainstem, innate defensive behaviours and physiological responses; (3) Cortical Nucleus (Co) — olfactory component.

Kluver-Bucy Syndrome

Result of removing medial temporal lobes (including hippocampus and amygdala) in rhesus monkeys. Characterised by a complete lack of fear — monkeys would approach stimuli that would ordinarily cause a fear response (e.g., a snake).

Mobbs et al. (2010) — tarantula fMRI study

Real tarantula moved toward participants' feet in an MRI scanner. As the spider got closer, both expected and experienced fear ratings increased. Brain imaging showed amygdala (left and right) and midbrain/dorsal ACC activated more when spider was close (compartments 4+5) vs. distant. When spider was distant, omPFC showed more activity.

Fear conditioning — definition and amygdala role

Fear conditioning = classical conditioning where the US is aversive. The amygdala is necessary for acquisition, storage, and expression of conditioned fear responses. Amygdala damage impairs conditioned fear in both animals and humans.

Implicit vs Explicit fear learning — amygdala vs hippocampus

Patient with amygdala damage: implicit fear learning (skin conductance response/SCR) is IMPAIRED; explicit knowledge (knowing blue square = shock) is INTACT. Patient with hippocampal damage: OPPOSITE — implicit fear learning is fine, but cannot explicitly state the CS-US relationship.

Instructed fear paradigm — findings

Participants told blue square = shock (shock never comes). Healthy controls show strong correlation between SCR (implicit) and amygdala activity. This correlation is absent in patients with amygdala damage — showing amygdala translates explicit threat knowledge into implicit physiological fear.

Amygdala and memory consolidation

Amygdala activity can enhance hippocampal memory consolidation for emotionally arousing events. In amygdala-lesioned rats, arousal does NOT improve memory consolidation. In humans, mild arousal increases memory for emotional events in healthy controls but not those with amygdala lesions.

Attentional blink — definition

During rapid serial visual presentation (RSVP), if a second target appears 150–450ms after the first, it often goes unnoticed. Smaller interval/lag = larger attentional blink. Highlights limits of the attentional system.

How does emotion affect the attentional blink?

Negative/emotional distractors increase the attentional blink at lag 2 (accuracy drops ~60–70% vs ~80% for neutral). At lag 8, all conditions recover to ~90%. Right temporal lobe resection patients show an even larger attentional blink for aversive and erotic distractors.

Why could Patient SM not recognise fear? — Adolphs eye gaze finding

Eye-tracking showed SM did not spontaneously look at the eye region of fearful faces (controls fixate heavily on eyes). When instructed to look at the eyes, SM could correctly recognise fear at levels matching controls (~85%). The amygdala normally directs attention to the eye region of fearful faces.

Subcortical fear signal — Whalen et al. eye whites finding

Fear is detectable from the amount of visible eye whites (scleras). Presenting only fearful eye whites (17ms) produced strong amygdala activation (~0.4). Happy eye whites produced minimal amygdala activation (~0.09). Eye blacks showed no difference between fearful and happy conditions.

Phelps et al. (2000) — race and amygdala, Experiment 1

White participants viewed Black and White faces in fMRI scanner. Implicit measures: Race IAT and startle eye-blink. Explicit measure: Modern Racism Scale. Result: Amygdala activation to Black faces positively correlated with IAT (r=.576) and startle (r=.556) but NOT with Modern Racism Scale (r=.047, ns). Suggests amygdala reflects culturally learned implicit associations, not explicit prejudice.

Phelps et al. (2000) — Experiment 2 (familiar faces)

When familiar, well-liked Black Americans (e.g. historical/celebrity figures) were used as stimuli: no difference in amygdala activation between Black vs. White faces, and no correlation with IAT or Modern Racism Scale. Amygdala activity reflects learned negative evaluation, not race per se.

Cunningham et al. (2004) — masked vs unmasked faces

Black faces presented for 30ms (masked) vs 525ms (unmasked). 30ms condition: significantly stronger amygdala activation to Black vs White faces, modulated by IAT. 525ms condition: Black > White faces in DLPFC/VLPFC/ACC (regions for monitoring and regulating racial bias) — interpreted as active suppression of negative evaluation.

Six factors that modulate race-related amygdala activity

(1) Task demand (Lieberman et al., 2005); (2) Race vs coalition (Van Bavel et al., 2008); (3) Eye gaze direction (Richeson et al., 2008); (4) Age/experience/exposure; (5) Novelty/salience (Telzer et al., 2013); (6) Skin-tone (Ronquillo et al., 2007). Also: pupil size (Demos et al., 2008) and trustworthiness (Engell et al., 2007) modulate amygdala activity.

Trustworthiness and amygdala (Engell et al., 2007)

Computer-generated untrustworthy faces produced greater amygdala activity than trustworthy faces. Both larger pupils AND more untrustworthy faces increased amygdala activity — the amygdala responds to social threat/anomaly signals.

Disgust and the insula

The insula is the key structure for disgust. It is connected to amygdala, medial PFC, and ACC, with reciprocal connections to attention, memory, and cognition regions. It is critical for interoception (perception of internal bodily states). Wacker et al. (2003): both feeling disgust and seeing others disgusted activate the same anterior insula region.

Why is the insula's role in disgust complicated?

The insula is functionally heterogeneous — activated by fear, disgust, anxiety, happiness, love, pain, bodily awareness, etc. MVPA (multivariate pattern analysis) techniques are needed to determine whether discrete neuronal populations within the insula specifically underlie disgust.

Reappraisal vs Suppression — definitions

Reappraisal (antecedent-focused): interpreting a potential emotion elicitor in a way that heightens or lessens its emotional impact — occurs BEFORE the response is generated. Suppression (response-focused): inhibiting or suppressing an emotional experience once it has already been activated or solidified.

Neural basis of reappraisal (Ochsner et al., 2004)

Reappraisal to increase OR decrease negative affect (NA) both recruit the DLPFC. Increasing NA specifically recruits rostral medial PFC. Decreasing NA specifically recruits right lateral PFC. Conclusion: cognitive control of emotion is largely prefrontally driven, with more recent work showing it requires PFC–subcortical interactions.

Three prominent emotion researchers and their positions (Adolphs, Barrett, Panksepp)

Ralph Adolphs (blue): views emotion as more modular, localised, domain-specific. Lisa Feldman Barrett (pink): views emotion as distributed, cortical, acquired, domain-general — supports constructionist account. Jaak Panksepp (gray): views emotion as innate, subcortical, domain-specific — supports locationist/evolutionary account.

Evolutionary view of emotion (Cosmides & Tooby)

Stimulus is detected as evolutionarily significant (e.g. predator threat) → fear emotion automatically activates a hardwired evolved program → all automatically: perception/attention shift, goal/motivation change, cognition (interpretive systems), physiology changes, behavioural decision rules. Information-gathering and learning systems are also activated.