1
Q
What is learning in neuroscience?
A
• Learning is a relatively lasting change in behaviour or brain function
• Caused by experience
• Occurs via changes in synaptic strength and neural circuits
2
Q
What is the biological basis of learning in the brain?
A
• Learning occurs through synaptic plasticity
• Involves strengthening or weakening of connections between neurones
• Alters prediction of outcomes
3
Q
What is associative learning?
A
• Learning that two events or actions are linked
• Forms associations such as stimulus–stimulus or action–outcome
• Core mechanism underlying conditioning
4
Q
What is Hebbian learning?
A
• Hebbian learning is activity-based learning
• Described as ‘neurones that fire together wire together’
• Repeated co-activation strengthens synapses
5
Q
How does Hebbian learning work at the synapse?
A
• Repeated firing of presynaptic and postsynaptic neurones together
• Leads to long-term potentiation (LTP)
• Increases synaptic efficiency
6
Q
Which receptors are central to Hebbian learning?
A
• NMDA receptors detect coincident activity
• Allow calcium influx
• Trigger insertion of AMPA receptors
7
Q
Which brain regions rely heavily on Hebbian learning?
A
• Hippocampus – memory formation
• Cerebral cortex – perceptual and skill learning
8
Q
Why is Hebbian learning NOT reward-based?
A
• It depends on neuronal co-activity, not outcomes
• No reward or punishment signal required
• Learning is correlation-based
9
Q
What is unsupervised learning?
A
• Learning occurs without external feedback
• The system detects patterns and regularities
• No teacher and no error signal
10
Q
What does the brain do during unsupervised learning?
A
• Groups similar inputs
• Extracts statistical structure
• Organises sensory information
11
Q
What are examples of unsupervised learning in humans?
A
• Face recognition
• Sensory development in infancy
• Perceptual organisation
12
Q
What is the neural basis of unsupervised learning?
A
• Hebbian plasticity
• Sensory cortex reorganisation
• Experience-dependent synaptic changes
13
Q
How is unsupervised learning relevant to psychiatry?
A
• Autism – atypical pattern detection
• Schizophrenia – abnormal salience assignment
14
Q
What is supervised learning?
A
• Learning guided by a teacher or error signal
• System is told what is correct or incorrect
• Learning aims to reduce error
15
Q
What is the key mechanism in supervised learning?
A
• Error correction
• Minimising difference between expected and actual outcome
16
Q
Which brain structures are associated with supervised learning?
A
• Cerebellum – motor learning
• Cortical feedback loops
17
Q
Why is supervised learning less central in psychiatry?
A
• Most psychiatric learning abnormalities involve reward and salience
• Reinforcement learning is more relevant clinically
18
Q
What is reinforcement learning?
A
• Learning driven by rewards and punishments
• Behaviour shaped by outcomes
• Central computational model in psychiatry
19
Q
What question does reinforcement learning answer?
A
• ‘Was the outcome better or worse than expected?’
• Behaviour is updated accordingly
20
Q
What is reward prediction error (RPE)?
A
• RPE = actual reward – expected reward
• Core signal that drives learning
21
Q
How does dopamine encode reward prediction error?
A
• Better than expected outcome → increased dopamine firing
• Worse than expected outcome → reduced dopamine firing
22
Q
Which brain systems mediate reinforcement learning?
A
• Dopamine neurones (VTA, substantia nigra)
• Basal ganglia
• Prefrontal cortex
23
Q
Why is dopamine NOT a pleasure signal?
A
• Dopamine codes prediction error
• Signals unexpectedness, not enjoyment
• Pleasure can occur without dopamine
24
Q
How is reinforcement learning altered in depression?
A
• Reduced reward sensitivity
• Blunted dopamine response
• Impaired reinforcement learning
25
**How is reinforcement learning altered in addiction?**
• Excessive learning of **drug-related cues**
• **Habit learning** dominates goal-directed control
• Pathological reinforcement
26
**How is reinforcement learning altered in schizophrenia?**
• **Aberrant salience**
• Dopamine mis-signals importance
• Faulty updating of beliefs and actions
27
**How is learning altered in obsessive–compulsive disorder?**
• Overactive **habit circuits**
• Impaired updating based on outcomes
• Behaviour persists despite negative consequences
28
**What is the key difference between Hebbian and reinforcement learning?**
• **Hebbian learning** = activity-based
• **Reinforcement learning** = outcome-based
• Reward is only involved in reinforcement learning
29
**What is a common exam trap involving dopamine?**
• Dopamine does **not code pleasure**
• Dopamine codes **prediction error**
30
**Which buzzword indicates Hebbian learning in exams?**
• **‘Fire together, wire together’**
31
**Which buzzword indicates reinforcement learning in exams?**
• **‘Prediction error’**
• **‘Better or worse than expected’**
32
**Which buzzword indicates unsupervised learning in exams?**
• **Pattern extraction**
• **Statistical regularities**
33
**What are the five exam-critical take-home points about learning?**
• Learning = **synaptic plasticity**
• Hebbian = **activity-based**
• Unsupervised = **pattern detection**
• Supervised = **error correction**
• Reinforcement = **dopamine and prediction error**
34
**What do we mean by learning in neuroscience?**
• **Learning** is a **relatively enduring change**
• Occurs in **behaviour**, **knowledge**, or **neural function**
• Caused by **experience**
• NOT due to **maturation**, **fatigue**, or **temporary states**
35
**Why is learning considered a biological process rather than just psychological?**
• Learning involves **physical changes in the brain**
• Includes changes in **synaptic strength**
• Alters **neuronal firing patterns**
• Reorganises **neural circuits**
• Represents **brain plasticity**
36
**What is synaptic plasticity?**
• **Synaptic plasticity** is the ability of a synapse to **increase or decrease its strength**
• Occurs in response to **neuronal activity**
• It is the **core cellular mechanism** of learning and memory
37
**What does activity-dependent plasticity mean?**
• Synapses change **only when neurones are active**
• Repeated activity strengthens connections
• Inactive synapses weaken (**‘use it or lose it’**)
• Explains learning and skill acquisition
38
**What is associative learning?**
• **Associative learning** is learning that **two events are linked**
• One event comes to **predict** another
• Forms the basis of **conditioning and habits**
39
**What happens in the brain during associative learning?**
• Neurones representing **Event A** and **Event B** fire together
• Repeated co-activation strengthens their synapses
• Stable neural representations are formed
• Explained by **Hebbian learning**
40
**What is Hebbian learning?**
• **Hebbian learning** is an activity-based learning rule
• Described as **‘neurones that fire together wire together’**
• Repeated co-activation strengthens synapses
• It is **NOT reward-based**
41
**What does Hebbian learning mean physiologically?**
• Repeated **synchronous firing** of pre- and postsynaptic neurones
• Leads to **long-lasting increases in synaptic efficiency**
• Synapse becomes faster and more reliable
• Expressed as **synaptic potentiation**
42
**What is long-term potentiation (LTP)?**
• **LTP** is a **persistent increase in synaptic strength**
• Follows repeated or strong synaptic activity
• Can last **hours to years**
• Input-specific and activity-dependent
43
**Why is LTP important for learning and memory?**
• LTP is the **best-studied biological mechanism** of learning
• Explains how experiences produce **lasting neural change**
• Strongly linked to memory formation
44
**Where was LTP first discovered and why is this important?**
• First identified in the **hippocampus (CA1 region)**
• Hippocampus is critical for **declarative and spatial memory**
• Supports link between **LTP and memory**
45
**Which neurotransmitter system is central to Hebbian learning?**
• **Glutamate** is the key neurotransmitter
• Acts at **AMPA** and **NMDA receptors**
• NMDA receptors are the critical **learning gate**
46
**Why are NMDA receptors crucial for Hebbian learning?**
• NMDA receptors act as **coincidence detectors**
• Require **glutamate binding** and **postsynaptic depolarisation**
• Ensure synapses strengthen only when neurones fire together
47
**What role does magnesium play at the NMDA receptor?**
• NMDA channel is blocked by **Mg²⁺ at rest**
• Postsynaptic depolarisation expels Mg²⁺
• Channel opens and **calcium enters**
• Makes NMDA receptors **voltage- and ligand-dependent**
48
**Why is calcium entry essential for learning?**
• **Calcium** is the intracellular learning signal
• Activates protein kinases (e.g. **CaMKII**)
• Triggers gene transcription
• Produces long-term synaptic change
49
**What happens to AMPA receptors during LTP?**
• **More AMPA receptors** are inserted into postsynaptic membrane
• Existing AMPA receptors become more conductive
• Same glutamate causes larger postsynaptic response
• Synapse is **potentiated**
50
**What structural synaptic changes occur with long-term Hebbian learning?**
• **Dendritic spines enlarge**
• Synapses stabilise
• New synapses may form
• Synaptic proteins are synthesised
51
**Which brain regions rely heavily on Hebbian learning?**
• **Hippocampus** – declarative and spatial memory
• **Neocortex** – long-term memory and skills
• **Sensory cortex** – experience-dependent tuning
• **Motor cortex** – motor learning
52
**Why is Hebbian learning considered unsupervised?**
• No **reward** is required
• No **punishment** is required
• No feedback or teacher signal
• Learning depends on **co-activation alone**
53
**What are key exam traps related to Hebbian learning?**
• It is **NOT dopamine-dependent**
• It is **NOT reward-based**
• It does **NOT require reinforcement**
• It is **NMDA-mediated and activity-based**
54
**Give a one-sentence exam-perfect summary of Hebbian learning.**
• **Hebbian learning** is an **activity-dependent, NMDA-mediated strengthening of synapses**, in which **repeated co-activation of neurones produces long-term potentiation and stable associative memory**
55
**What are computational learning models in neuroscience?**
• **Computational learning models** describe the **rules** the brain uses to learn from experience
• Explain **how synapses change**
• Explain **how predictions are updated**
• Help understand **normal learning** and **psychiatric disorders**
56
**How many core types of computational learning models does the brain use?**
• The brain uses **THREE core learning types**
• **Unsupervised learning**
• **Supervised learning**
• **Reinforcement learning**
• Each relies on **different brain circuits** and fails in **different psychiatric illnesses**
57
**What is unsupervised learning?**
• **Unsupervised learning** occurs **without reward**
• Occurs **without punishment**
• Occurs **without feedback or a teacher**
• The brain learns by detecting **patterns**
58
**What question does unsupervised learning answer?**
• Unsupervised learning answers:
• **‘What things belong together?’**
• Helps discover **structure**, **categories**, and **meaning**
• No concept of right or wrong
59
**Give simple real-life examples of unsupervised learning.**
• **Face recognition**
• **Speech sound learning** in infancy
• **Object recognition**
• **Perceptual grouping**
• Occurs **automatically**, without feedback
60
**What is the biological mechanism behind unsupervised learning?**
• Main mechanism is **Hebbian learning**
• Repeated **co-activation** strengthens synapses
• Builds stable neural representations
• Occurs **without dopamine** and **without reward**
61
**Which brain areas mainly use unsupervised learning?**
• **Sensory cortices** (visual, auditory, somatosensory)
• **Association cortices** (temporal and parietal)
• These areas extract **patterns from raw input**
62
**Why is unsupervised learning especially important in early development?**
• Infants receive **no explicit teaching**
• Learning occurs through **exposure alone**
• Drives **language**, **face recognition**, and **social cue detection**
• Strongest during **critical periods**
63
**How is unsupervised learning linked to autism?**
• Pattern detection is **atypical**
• Focus on **local details**, not global meaning
• Leads to **sensory hypersensitivity**
• Difficulty interpreting **social signals**
• Brain extracts **different patterns** from the same environment
64
**How is unsupervised learning linked to schizophrenia?**
• Pattern extraction becomes **noisy**
• Unrelated events are grouped together
• Causes **false associations**
• Leads to **ideas of reference** and **delusions**
• Known as **aberrant salience**
65
**What is a key exam trap about unsupervised learning?**
• It is **NOT random learning**
• It is **NOT reward-based**
• It is **statistical pattern learning**
66
**What is supervised learning?**
• **Supervised learning** involves an **explicit error signal**
• A ‘teacher’ indicates **right or wrong**
• Learning occurs by **correcting mistakes**
67
**What question does supervised learning answer?**
• Supervised learning answers:
• **‘How can I improve performance toward a known goal?’**
• Focuses on **accuracy** and **fine adjustment**
68
**What is the key biological feature of supervised learning?**
• **Error correction**
• Difference between **expected** and **actual outcome**
• Synapses change to **reduce future errors**
69
**Which brain structure is most important for supervised learning?**
• The **cerebellum**
• Compares intended with actual movement
• Uses error signals to fine-tune output
• Damage causes **poor motor learning** and **ataxia**
70
**Why is supervised learning less important in psychiatry?**
• Psychiatric symptoms involve **motivation**, **meaning**, and **salience**
• These rely more on **reinforcement learning**
• Error correction plays a smaller role
71
**What is reinforcement learning?**
• **Reinforcement learning** is driven by **rewards and punishments**
• Outcomes are evaluated as **better or worse than expected**
• The brain asks **‘Was that outcome worth it?’**
72
**What is reward prediction error (RPE)?**
• **Reward prediction error** is the difference between:
• **Expected outcome**
• **Actual outcome**
• Formula: **Actual − Expected**
73
**How does dopamine relate to reward prediction error?**
• Dopamine firing **increases** if outcome is **better than expected**
• Dopamine firing **decreases** if outcome is **worse than expected**
• Dopamine codes **learning signals**, NOT pleasure
74
**Which brain circuits are involved in reinforcement learning?**
• **VTA and substantia nigra** (dopamine)
• **Basal ganglia** (actions and habits)
• **Prefrontal cortex** (decision-making)
• **Limbic system** (emotional value)
75
**How is reinforcement learning abnormal in depression?**
• Dopamine responses are **blunted**
• Rewards fail to update behaviour
• Leads to **anhedonia**, **apathy**, and **low motivation**
• Rewards stop ‘teaching’ the brain
76
**How is reinforcement learning abnormal in addiction?**
• Drugs cause **excessive dopamine release**
• Rewards are **over-valued**
• Drug cues become **pathologically reinforced**
• Leads to **craving**, **compulsion**, and **habitual use**
77
**How is reinforcement learning abnormal in schizophrenia?**
• Dopamine signals occur at **inappropriate times**
• Neutral events feel **important**
• Causes **aberrant salience**
• Leads to **delusions and psychosis**
78
**How is reinforcement learning abnormal in OCD?**
• **Habit circuits** dominate behaviour
• Outcome updating is impaired
• Leads to **repetitive compulsions**
• Behaviour persists despite insight
79
**What are the biggest exam traps in computational learning models?**
• **Hebbian ≠ reward**
• **Dopamine ≠ pleasure**
• **Unsupervised ≠ random**
• **Reinforcement ≠ simple reward**
80
**Give a simple mnemonic for learning models.**
• **HUR Learning**
• **H**ebbian → **H**appens together
• **U**nsupervised → **U**nlabelled patterns
• **R**einforcement → **R**eward prediction error
81
**Why are computational learning models important for understanding psychiatric disorders?**
• Many psychiatric symptoms arise from **abnormal learning**
• Involve **faulty belief updating**
• Involve incorrect assignment of **salience**
• Biologically reflect **synaptic plasticity abnormalities**
• Involve **dopamine signalling errors**
• Occur in **cortico–striatal circuit dysfunction**
• Psychiatric illness = **disordered learning**, not just chemical imbalance
82
**What is the shared pathophysiological theme across psychiatric learning disorders?**
• Failure of **pattern formation** (unsupervised learning)
• Failure of **error correction**
• Failure of **reward prediction error signalling**
• Failure to switch between **goal-directed and habitual control**
• Deficits occur in **specific neural circuits**, not globally
83
**How does unsupervised learning normally work in the brain?**
• Repeated **co-activation** of sensory neurones
• **Hebbian plasticity** strengthens correct associations
• Stable **internal models** of the world form
• Depends on **NMDA-mediated plasticity**
• Requires balanced **excitation and inhibition**
• Requires low cortical **noise**
84
**What goes wrong with unsupervised learning in autism?**
• Altered **synaptic plasticity**
• **Excitation–inhibition imbalance** (↑ glutamate / ↓ GABA)
• Reduced **integration across cortical areas**
• Over-learning of **local features**
• Under-learning of **global patterns**
• Sensory cortex plasticity is **over-detailed but poorly integrated**
85
**How does abnormal unsupervised learning explain autistic symptoms biologically?**
• Sensory inputs encoded **too precisely**
• Reduced **noise filtering**
• Impaired extraction of **higher-order meaning**
• Produces **sensory hypersensitivity**
• Causes difficulty with **social pattern recognition**
• World feels **overwhelming and fragmented**
86
**What goes wrong with unsupervised learning in schizophrenia?**
• Cortical learning becomes **unstable**
• **Random co-activation** is reinforced
• Reduced **signal-to-noise ratio**
• NMDA receptor **hypofunction**
• Cortical **disinhibition**
• Excess background **dopamine influence**
87
**How do unsupervised learning failures lead to delusions and perceptual abnormalities?**
• Unrelated stimuli are **grouped together**
• **False associations** are learned
• Internal representations lose **boundaries**
• Leads to **ideas of reference**
• Produces **delusions and perceptual distortions**
• Known as **aberrant pattern formation**
88
**How does supervised learning normally work biologically?**
• Relies on explicit **error signals**
• Compares **intended vs actual outcomes**
• Uses **cerebellar error correction**
• Synapses adjust to **reduce future error**
• Improves **precision and timing**
89
**What happens when supervised learning systems fail?**
• **Poor error correction**
• Inaccurate predictions
• **Inflexible or clumsy behaviour**
• Seen in **motor coordination problems**
• Seen in **speech and timing abnormalities**
90
**Why is supervised learning less central to core psychiatric symptoms?**
• Psychiatric disorders involve **meaning**
• Involve **motivation and value**
• Involve **belief formation**
• These rely on **reinforcement learning**
• Depend on **dopamine-based cortico-striatal circuits**
91
**What is the normal biological role of dopamine in reinforcement learning?**
• Dopamine signals **reward prediction error (RPE)**
• ↑ firing = outcome **better than expected**
• ↓ firing = outcome **worse than expected**
• Updates **synaptic weights**
• Guides **future decisions**
• Shapes **habits and motivation**
92
**What goes wrong with reinforcement learning in depression?**
• Dopamine **RPE signals are blunted**
• Positive outcomes fail to **reinforce behaviour**
• Negative outcomes are **overweighted**
• Reduced **dopaminergic firing**
• Reduced **ventral striatal responsiveness**
• Excess **top-down inhibitory control**
93
**How do reinforcement learning abnormalities produce depressive symptoms?**
• Rewards no longer feel **informative**
• Behaviour stops being **reinforced**
• **Motivation collapses**
• Leads to **anhedonia**
• Leads to **apathy**
• Produces **learned helplessness**
• Core pathology = **failure of positive learning**
94
**What goes wrong with reinforcement learning in addiction?**
• Drugs cause **massive non-physiological dopamine release**
• **RPE signals are exaggerated**
• Learning becomes **distorted**
• Drug-cue associations are **over-strengthened**
• Natural rewards lose **learning value**
95
**How does addiction shift behaviour biologically?**
• Shift from **goal-directed learning** (ventral striatum)
• Shift toward **habit-based learning** (dorsal striatum)
• Produces **compulsion**
• Causes **loss of control**
• Behaviour persists **despite harm**
• Addiction = **pathological reinforcement learning**
96
**What goes wrong with reinforcement learning in schizophrenia?**
• Dopamine fires at **inappropriate times**
• **RPE signals** occur without real rewards
• Dysregulated **dopamine synthesis**
• Impaired **cortical control** of midbrain dopamine
97
**How do reinforcement learning abnormalities cause psychotic symptoms?**
• Neutral events feel **important**
• Meaning is assigned **incorrectly**
• Beliefs updated using **false signals**
• Leads to **aberrant salience**
• Produces **delusions and psychosis**
• Learning occurs from **noise instead of signal**
98
**What goes wrong with reinforcement learning in OCD?**
• **Habit learning** dominates behaviour
• Outcome evaluation is **impaired**
• **Extinction learning** is weak
• Overactive **cortico-striato-thalamo-cortical loops**
• Reduced **prefrontal inhibition of habits**
99
**How do reinforcement learning abnormalities explain compulsions?**
• Behaviour no longer guided by **outcomes**
• **Habits run automatically**
• Anxiety relief acts as **negative reinforcement**
• Produces **repetitive behaviours**
• Behaviour persists **despite insight**
100
**Give a one-sentence pathophysiological summary for psychiatry exams.**
• Psychiatric disorders arise from **disrupted synaptic plasticity and dopamine-mediated learning signals**
• Leads to **faulty pattern formation**
• Causes **abnormal salience**
• Produces **impaired reward learning**
• Allows **maladaptive habits to dominate behaviour**
Cranial nerves
flashcards
Decks in class (64)
# Cards
-
CN3 OCCULOMOTOR55
-
CN4 TROCHLEAR51
-
Trigemnial nerve154
-
FACIAL NERVE75
-
VESTIBULOCOCHLEAR NERVE94
-
GLASSOPHARYNGEAL50
-
VAGAS78
-
ACCESSORY59
-
Hypoglossal84
-
Midbrain65
-
PONS80
-
Medulla151
-
Cerebellum100
-
Dexcending pyrMIDl trCT17
-
Cerebellar tract95
-
Cerebellar cells44
-
Efferent Cerebellar tract59
-
Cerebellar lesions92
-
Spinal chord anatomy132
-
Cerbrovascular circ151
-
CNS Cells236
-
Cerebral neocortex cells249
-
Bloodbrain barrier29
-
CSF system55
-
Action potential and synapse115
-
white matter tracts71
-
Cerbral cortex layers225
-
Synapses47
-
plasticity etc83
-
meurogenesis and migration0
-
Cortisol + CCK35
-
Ghrehlin/Leptin103
-
Endocannabinoids31
-
Endokephalin/Endorphins73
-
Orexin49
-
3.3.6 Links between neurotransmitter systems and findings from genetic association studies in psychiatry.178
-
high yeild31
-
neural circuit - appetite175
-
neural circuit - thirst72
-
neural circuit - aggression153
-
neural circuit - pain111
-
learning100
-
Habit and compulsions148
-
Motivation, reward and pelasure147
-
emotional regulation109
-
perception135
-
executive function + empathy/tom211
-
DMN and Salience118
-
Basic concepts: chromosomes, cell division, gene structure, transcription and translation,19
-
structure of the human genome, patterns of inheritance.29
-
Neurodevelopmental models40
-
neurodev model88
-
neurobiology of attatchment102
-
neurodev model88
-
stress182
-
CBT128
-
attatchment53
-
Learning and therapy41
-
neurobio learning39
-
LEARNING THERAPY40
-
LEARNING41
-
Brain injury86
-
Alzheimer's58
-
Vascular0
