MCY204C21

COGNITIVE NEUROSCIENCE II
Comprehensive Exam Notes

Units I–V: Language • Reasoning • Cognitive Control • Social Cognition • Consciousness

MCY204C21: Cognitive Neuroscience II | Page 1

UNIT I: NEUROSCIENCE OF LANGUAGE
1.1 Anatomy of Language
Most language functions are controlled by the left hemisphere of the brain. Studies of split-brain
patients and people with brain damage have clearly demonstrated this lateralization. The critical
language areas lie around the Sylvian fissure, forming the left perisylvian language network.

Key Language Areas

•​ Wernicke's Area: Located in the posterior superior temporal gyrus of the left temporal lobe.
Responsible for understanding spoken and written language. Damage produces fluent but
meaningless speech with poor comprehension.
•​ Broca's Area: Located in the pars triangularis and pars opercularis of the left inferior frontal
gyrus. Controls speech production, grammar, and sentence formation. Damage produces
slow, non-fluent, effortful speech.
•​ Angular Gyrus: Part of the inferior parietal lobe; critical for reading and writing. Damage
causes alexia (reading disorder) and agraphia (writing disorder).
•​ Supramarginal Gyrus: Connects speech sounds to meaning. Damage causes difficulty
linking sounds with meaning.
•​ Anterior Temporal Cortex: Responsible for semantic processing—understanding word and
sentence meaning. Damage impairs comprehension.
•​ Left Insular Cortex: Involved in speech coordination and articulation. Damage causes
apraxia of speech.
•​ Arcuate Fasciculus: White matter tract connecting Wernicke's area to Broca's area. Damage
produces conduction aphasia.

Motor Areas Involved in Speech

•​ Premotor Cortex: Plans speech movements. Translates Broca's area plans into motor
sequences for lips, tongue, jaw, and vocal cords. Damage → apraxia of speech.
•​ Primary Motor Cortex: Executes speech movements. Sends direct commands to speech
muscles. Damage → dysarthria.
•​ Supplementary Motor Area (SMA): Initiates speech and maintains fluency of internally
generated speech. Damage → reduced spontaneous speech.
•​ Basal Ganglia: Regulates timing, rhythm, and sequencing of speech. Damage (e.g.,
Parkinson's disease) → slow, monotone, reduced-volume speech.
•​ Cerebellum: Coordinates timing, precision, and error correction. Damage → ataxic speech
with irregular rhythm.
•​ Thalamus: Acts as relay center connecting cortical and subcortical language areas. Damage
→ disrupted fluency and word retrieval.

Role of the Right Hemisphere

The right hemisphere contributes to language in specific ways: the right superior temporal sulcus
processes prosody (rhythm and intonation), and the right prefrontal cortex, middle temporal gyrus,
and posterior cingulate activate when processing metaphorical meaning in sentences.

MCY204C21: Cognitive Neuroscience II | Page 2

1.2 Brain Damage and Language Deficits — Aphasias
Aphasia is a broad term for impairments in language comprehension and production resulting from
neurological damage, even when articulatory mechanisms are intact. Aphasia affects approximately
40% of all stroke patients and may be accompanied by dysarthria (loss of articulatory muscle
control) or speech apraxia (deficit in motor planning of speech movements).

Broca's Aphasia (Non-fluent / Expressive Aphasia)

•​ First described by: Paul Broca, 19th-century Parisian physician, following autopsy of his
patient 'Tan.'
•​ Lesion site: Posterior left inferior frontal gyrus (pars triangularis + pars opercularis), plus
deeper regions including insular cortex and basal ganglia.
•​ Key features: Slow, broken, effortful telegraphic speech; omission of function words (is, the,
and); difficulty with grammar (agrammatism); anomia (word-finding difficulty); articulation
difficulties.
•​ Comprehension: Intact for simple sentences but impaired for complex grammatical structures
(e.g., passive voice).
•​ Modern understanding: Broca's area alone does not cause the full syndrome. Nina Dronkers
showed that many patients with Broca's area lesions do NOT show Broca's aphasia; it results

📌
from damage to a network of brain areas.
Wernicke proposed that Broca's area stores 'motor memories' — the exact sequence of
muscle movements needed to pronounce words.

Wernicke's Aphasia (Receptive / Posterior Aphasia)

•​ First described by: Carl Wernicke, German physician.
•​ Lesion site: Posterior superior temporal gyrus (Wernicke's area) plus nearby posterior
temporal lobe and white matter connections.
•​ Key features: Severely impaired comprehension of both spoken and written language; fluent
speech with normal rate, rhythm, and intonation, but content is nonsensical; jargon aphasia;
paraphasias (substituting wrong words).
•​ Important note: Damage to Wernicke's area alone does not always cause long-lasting
problems; severe and persistent Wernicke's aphasia requires damage to the surrounding
posterior temporal regions as well.

Conduction Aphasia
•​ Anatomical basis: Damage to the arcuate fasciculus — the white matter pathway connecting
Wernicke's area to Broca's area (proposed by Lichtheim, 1880s).
•​ Key features: Good comprehension; fluent speech; severe difficulty repeating words and
sentences; patient hears own speech errors but CANNOT correct them; phonemic
paraphasias (wrong sound substitutions).
•​ Why it happens: Wernicke's area (understanding) and Broca's area (production) are both
intact, but the connection between them is damaged. Recognized spoken word cannot
smoothly transfer to speech production.
•​ Modern view: Norman Geschwind emphasized that language is supported by networks;
damage to the insula or nearby auditory areas can also cause similar problems.

MCY204C21: Cognitive Neuroscience II | Page 3

Other Aphasias (Lichtheim's Model)
•​ Transcortical Sensory Aphasia: Damage between concept area and Wernicke's area.
Patients CAN repeat but cannot understand. Usually involves supramarginal and angular gyri.
•​ Transcortical Motor Aphasia: Damage between concept area and Broca's area. Patients
CAN repeat and understand but cannot initiate spontaneous speech; may show echolalia.
•​ Global Aphasia: Damage to both Broca's and Wernicke's areas plus surrounding tissue.
Severe impairment of all language functions.

1.3 Language in the Human Brain: Words and Representations

Model of the Mental Lexicon
The mental lexicon is our internal dictionary. It is organized semantically — words are linked by
meaning into networks. Related concepts activate each other (e.g., 'dog' activates 'animal,' 'cat,'
'pet').
•​ Lemmas: Abstract lexical entries containing a word's grammatical category, gender, meaning,
and context of use — but NOT its sound form. Lemmas are the building blocks of sentence
structure.
•​ Word forms (Lexemes): The phonological (sound) and morphological structure of the word.
Activated after the lemma is selected.
•​ Tip-of-the-Tongue (TOT) state: Occurs when lemma is available (meaning known) but sound
form is temporarily unavailable. Demonstrates that meaning and sound are stored separately.

1.4 Language Comprehension

Understanding Spoken Input (Speech Perception)
Speech comprehension begins with the auditory cortex processing incoming sounds, then
proceeding through temporal lobe regions to extract phonological, lexical, and semantic information.
The perisylvian network (including Wernicke's area and the superior temporal sulcus) is essential
for extracting meaning from continuous speech.

Written Input: Reading Words

Reading involves a separate neural pathway from speech comprehension. The occipitotemporal
region, specifically the Visual Word Form Area (VWFA) in the left fusiform gyrus, is specialized for
processing written symbols.
•​ Visual Word Form Area (VWFA): Also called the 'word box.' Located in the left
occipitotemporal sulcus. Responds specifically to written symbols (letters, words) but not faces
or textures. Active regardless of writing system (e.g., Japanese kana/kanji and alphabetic
scripts).
•​ ERP evidence: Nobre et al. (1994) found a large negative potential at ~200ms in
occipitotemporal regions in response to letter strings but not other stimuli. This region was
insensitive to lexical or semantic features.

MCY204C21: Cognitive Neuroscience II | Page 4

 •​ Cohen et al. (2000): Using fMRI and ERP, showed that after VWFA processing, further word
processing is the same regardless of which visual field the word first appeared in, confirming

📌
convergence in the left hemisphere.
Neural Substrates of Written-Word Processing involve V1 → occipital cortex → VWFA →
Wernicke's area → semantic processing → Broca's area for phonological output.

1.5 Neural Models of Speech Production

Levelt's Model (1989)
Willem Levelt proposed a serial, modular model of speech production with distinct, sequentially
processed stages:
•​ Stage 1 — Macroplanning: Deciding WHAT to say (the message's goal and content).
•​ Stage 2 — Microplanning: Deciding HOW to say it (specific words, sentence structure,
perspective).
•​ Stage 3 — Grammatical Encoding: Building the surface structure of the sentence (subject,
object, word order). Involves lemma selection.
•​ Stage 4 — Morphological Encoding: Adding grammatical endings (e.g., goat → goats, run →
running).
•​ Stage 5 — Phonological Encoding: Activating the sound structure and metrical pattern
(syllables, stress).
•​ Stage 6 — Articulation: Motor patterns sent to tongue, lips, mouth, vocal cords for physical

📌
speech production.
Levelt's model is MODULAR — each stage is independent, with no feedback from later to
earlier stages. Processing is strictly one-way.

Dell's Interactive Model (1986)

Gary Dell proposed that speech production stages OVERLAP and INTERACT. Unlike Levelt's serial
model, phonological activation begins as soon as meaning and grammar start activating — before
earlier stages are complete. Critically, the model allows FEEDBACK: sound activations can
influence meaning and grammar. This is a two-way interactive system.

Sahin et al. (2009) — Neural Basis of Word Production

Using intracranial electrodes in Broca's area, researchers found a triphasic electrical response:
•​ Wave 1 (~200 ms): Lexical identification — accessing the word from the mental lexicon.
•​ Wave 2 (~320 ms): Grammatical processing — applying inflectional forms (tense, number).
Influenced by inflectional demands, NOT phonology.
•​ Wave 3 (~450 ms): Phonological encoding — organizing sounds for articulation.
•​ Motor preparation at ~500-550 ms; actual speech at ~600 ms.
These findings support serial processing in Broca's area and show that lexical, grammatical, and
phonological circuits are spatially (though closely) separated within Broca's area.

Brain Regions Activated During Speech Production (Imaging Studies)

•​ Inferior temporal cortex — Lexical retrieval (word meaning during picture naming)

MCY204C21: Cognitive Neuroscience II | Page 5

•​ Broca's area / Frontal operculum — Phonological encoding
•​ Primary motor cortex (bilateral) — Execution of speech movements
•​ Supplementary Motor Area (SMA) — Motor planning and coordination
•​ Insula — Fine motor coordination for articulation
•​ Basal ganglia — Motor sequencing and initiation
•​ Thalamus — Relay and coordination of motor signals
•​ Cerebellum — Timing, precision, and motor coordination

MCY204C21: Cognitive Neuroscience II | Page 6

UNIT II: NEUROSCIENCE OF REASONING AND
PROBLEM SOLVING
2.1 Introduction: Reasoning vs Problem Solving
Reasoning and problem solving are central to cognition, embodying what we call thinking and
intelligence. René Descartes' 'I think, therefore I am' reflects the centrality of reasoning to human
existence. Both share similarities (planning, foresight, logic, goal-orientation) but differ in their
processes.
•​ Reasoning: Involves logical inferences based on well-defined rules. Follows a structured
logical pattern. Conclusions can be evaluated as strictly true or false. Example: Aristotle's
syllogism.
•​ Problem Solving: Involves moving from a starting point to a desired goal without clearly
defined intermediate steps. Rules may not be specified; multiple solutions may exist. Requires
creative and associative thinking.

2.2 Deductive Reasoning

Definition and Syllogisms
Deductive reasoning involves argument from premises to a specific, logically guaranteed
conclusion. The truth of the conclusion must follow solely from the premises, regardless of whether
those premises are true in real life. The classic form is the syllogism: two premises + one
conclusion.
•​ Valid syllogism example: 'All humans are mortal. Socrates is human. Therefore, Socrates is
mortal.'
•​ Invalid syllogism example: 'All cats are animals. All dogs are animals. Therefore, all dogs are
cats.' — This fails because the conclusion does not logically follow.

Valid Inference Forms

•​ Modus Ponens (VALID): 'If P, then Q. P. Therefore, Q.' — The mode of argument that affirms.
The safest form of deductive reasoning.
•​ Modus Tollens (VALID): 'If P, then Q. Not Q. Therefore, not P.' — The mode that denies.
People often have difficulty with this form, especially in abstract contexts.

Common Logical Errors

•​ Denying the Antecedent (INVALID): 'If P, then Q. Not P. Therefore, not Q.' — Error: Q may
still occur for other reasons. E.g., 'If it rains, the ground gets wet. It is not raining. Therefore,
the ground is not wet.' — WRONG, ground may be wet from sprinklers.
•​ Affirming the Consequent (INVALID): 'If P, then Q. Q. Therefore, P.' — Error: Q could have
happened for another reason. E.g., 'If she is anxious, her heart rate increases. Her heart rate
increased. Therefore, she is anxious.' — WRONG, exercise could cause the same.

MCY204C21: Cognitive Neuroscience II | Page 7

Context Effects on Reasoning (Wason Card Task)
Cosmides and Tooby demonstrated that reasoning performance improves dramatically when
problems are framed in social contexts (e.g., permission rules involving cheating detection)
compared to abstract contexts. fMRI studies show that social reasoning activates additional right
prefrontal and parietal regions beyond those used for abstract reasoning.

2.3 Theories of Deductive Reasoning

A) Mental Logic Theory
Humans possess an internal system of mental logic rules. When reasoning, we abstract the
problem structure and apply logical rules step by step. This view assumes reasoning is abstract,
rule-governed, and similar to formal logic.
•​ Limitation: People do NOT always reason abstractly. They use prior knowledge (belief bias)
and struggle when premises contradict real-world knowledge (e.g., 'No cigarettes are
harmful'). The art gallery example (paintings on one vs. multiple walls) shows that pure rule
application can fail when context and spatial modeling matter.

B) Mental Model Theory (Philip Johnson-Laird)

People do not use abstract logical rules. Instead, they construct one or more mental representations
(models) of the situation, assuming the premises are true, then evaluate whether the conclusion fits
all possible models.
•​ If they CANNOT construct a model where the conclusion is false → they accept the
conclusion as true.
•​ People are good at hypothesis-confirming (verifying what seems possible) but poor at
hypothesis-falsifying (checking if the conclusion could be false).

📌
•​ Errors occur because people stop after the first model without searching for alternatives.
The mental model approach explains why deductive reasoning resembles inductive
reasoning at a neural level — both rely on mental simulation and experience-based
understanding, not formal logic.

2.4 Neural Systems Supporting Deductive Reasoning

•​ Frontal Lobes (Critical): Deductive reasoning requires functionally intact frontal lobes,
especially the lateral prefrontal cortex (LPFC).
•​ Left Frontal Lobe: Primarily supports language-based, explicit reasoning. Research by Vinod
Goel: deduction activates inferior parts of the left frontal lobe, including Broca's area.
•​ Left-Brain Interpreter: Michael Gazzaniga's concept — the left hemisphere forms inferences
and associations among multiple concepts. Patients with left frontal lobe damage show normal
basic reasoning but do not improve with social information (unlike healthy controls).
•​ Right Hemisphere: May also be involved, in regions homologous to Broca's area, under
certain types of deduction. Conditions not yet fully established.
•​ Temporal Lobe: Involved when reasoning uses well-known semantic categories (e.g.,
'doctors,' 'animals'). Left temporal lobe damage → broad impairment in syllogistic reasoning;
right temporal lobe damage → impaired reasoning requiring content manipulation.
•​ Parietal Cortex: Supports reasoning about arbitrary categories.
•​ Hippocampus: Contributes to spatial reasoning.

MCY204C21: Cognitive Neuroscience II | Page 8

 •​ Right Frontal Lobe: Damage impairs: (1) changing conclusion form to match premises, (2)
multiple relations ('X larger than Y, Y larger than Z, which smallest?'), (3) reaching true
conclusions from false premises.

2.5 Inductive Reasoning

Inductive reasoning involves determining the likely truth of a conclusion from probabilistic, often
imperfect premises. Unlike deduction, the conclusion is probable but NOT guaranteed. Truth cannot
be derived from premises alone; prior empirical knowledge is required.

Types of Inductive Arguments

•​ Argument from Analogy: Inferring something about X because it is similar to Y. Example:
'House cats purr when happy. Lions are related to house cats. Therefore, lions may purr when
happy.'
•​ Reasoning from Category Probability: Using category membership. Example: 'Most birds
can fly. A magpie is a bird. Therefore, a magpie can probably fly.'
•​ Generalization from Examples: Inferring about a whole group from specific instances.
Example: 'Mary is CIA. Mary has high security clearance. Therefore, all CIA employees have
high security clearance.' — RISKY because it generalizes from limited examples.

Factors Affecting Inductive Reasoning

Inductive reasoning is more difficult when: events are separated in time; an event has multiple
causes; causes/relations are abstract; category shifts are required; or conclusions contradict prior
beliefs. Children use salient features for induction (wood chip floats because it's 'light'); adults use
heuristics that can cause biased judgments.

2.6 Neural Systems Supporting Inductive Reasoning

•​ Left Prefrontal Cortex: Critical for inductive reasoning. Involved in both language processing
and executive functions. PET and fMRI consistently show left PFC activation during inductive
tasks.
•​ Dorsolateral PFC (DLPFC): Damage → perseveration (continuing rules no longer
applicable); difficulty learning new rules from examples/feedback. Impairs Tower of Hanoi
performance through deficits in planning and flexible thinking.
•​ Left Lateral PFC: Critical for generating NEW hypotheses — essential for induction but NOT
needed for deduction.
•​ Induction vs. Deduction (Goel & Dolan): Both activate left PFC but in DIFFERENT
sub-regions: Induction → dorsolateral PFC (hypothesis generation and testing); Deduction →
ventrolateral PFC (language and syntax).
•​ Important Case Study: A patient with damage to inferior frontal lobe + temporal lobe (Broca's
area) was severely agrammatical — could not form sentences or match sentences to pictures
grammatically. Yet his DLPFC was intact (good Wisconsin Card Sorting Test), and he could
perform inductive reasoning when tasks were presented aurally or pictorially. This shows
inductive reasoning can operate INDEPENDENTLY of language abilities.
•​ Basal Ganglia: Also important for inductive reasoning, particularly for learning from examples.
Parkinson's disease patients (basal ganglia affected) show difficulty forming associations.

MCY204C21: Cognitive Neuroscience II | Page 9

 •​ Caudate Nucleus: Body and tail activate as categories are learned; head activates when
feedback is most informative. Contributes to learning, feedback processing, and category
formation.

2.7 Problem Solving

Problem solving involves finding a solution to a goal when the method is not immediately obvious.
Classic example: Archimedes and the golden crown — discovering that density determines water
displacement (Eureka!). Key components: a goal, an undefined set of steps, and a large problem
space (set of possible ways to approach the problem).

Problem Space
The set of all possible ways a problem can be approached. Large problem spaces (like early chess)
require heuristics; small problem spaces (chess endgame with few pieces) require logical reasoning
rather than problem solving.

2.8 Heuristics in Problem Solving

When faced with complex problem spaces, people use heuristics — simple mental shortcuts that
are practical but not always optimal.
•​ Hill-Climbing Heuristic: Taking the step that minimizes the perceived difference between
current state and goal. Useful for reducing complexity but fails when problems require
backtracking. Example: chess sacrifice.
•​ Repeated Application of a Successful Rule: Demonstrated in Luchins' Water Jar Problems:
after solving several similar problems using one method, people continue using it even when a
simpler solution exists.
•​ Functional Fixedness: The inability to see alternative representations of a problem; fixating
on one conception at the cost of missing simpler solutions. Luchins' studies: >80% continued
the complex formula even when a trivial solution (23 - 3 = 20) was available. Also observed in
everyday anagram-solving tasks.

2.9 Insight in Problem Solving

Insight is the sudden realization of a solution after reaching an impasse, typically accompanied by a
'Eureka' feeling. Usually occurs after trial-and-error and heuristic approaches have failed.
•​ Reorganization Theory (Wolfgang Köhler): Insight occurs when one way of viewing the
problem is replaced by a new, more promising perspective.
•​ Warmth Ratings: People rate their closeness to a solution as very low even moments before
insight — yet implicit semantic priming effects show the brain processes the solution
unconsciously BEFORE the person becomes aware. Insight feels sudden at a conscious level
but the underlying process is gradual.
•​ Duncker Candle Problem: Box of tacks + candle + matches. Solution: use box as a platform.
People fail due to functional fixedness (seeing box only as container).
•​ Rope Problem: Two ropes too far apart to hold simultaneously. Solution: use flashlight as a
pendulum weight.

2.10 Neural Systems Supporting Problem Solving

MCY204C21: Cognitive Neuroscience II | Page 10

•​ Frontal Lobes: Critical for planning, executive control, and structured problem solving (e.g.,
Tower of Hanoi). Damage → poor performance on planning tasks.
•​ Right Hemisphere: Associated with insight in neuroimaging studies. Implicit associations and
priming effects stronger when cues presented to left visual field (right hemisphere).
•​ Anterior Temporal Cortex: Activated when subjects solve problems with insight. Involved in
semantic processing and forming associations. EEG shows gamma wave burst in anterior
temporal cortex about 300ms before conscious insight report.
•​ Lateral Prefrontal Cortex: Helps establish links between stimuli and responses; can be
beneficial (paring problem space) or detrimental (eliminating correct solutions). Patients with
lateral PFC damage: worse on standard problems but BETTER on problems requiring
breaking conventional rules.
•​ Cingulate Cortex, Posterior Parietal, Medial Temporal Lobe: Active for more complex
insight tasks (riddles) alongside executive functions.
•​ Role of Frontopolar Cortex: Associated with integrating abstract information from different
sources; contributes to broad behavioral strategies; activated when people must compare two

📌
internally generated judgments.
Insight produces a NEGATIVE ERP (event-related potential) at ~400ms at midline frontal
sites when the insight solution is presented, reflecting rapid meaningful processing.

MCY204C21: Cognitive Neuroscience II | Page 11

UNIT III: COGNITIVE CONTROL AND DECISION MAKING
3.1 Meaning of Cognitive Control
Cognitive control (also called executive function) refers to mental processes that allow us to
regulate thoughts and actions according to goals. It enables us to choose appropriate actions from
many possible responses, override automatic or habitual behaviors, think flexibly, plan actions,
simulate outcomes, solve problems, and make goal-directed decisions.
Example: Wanting to eat a doughnut but choosing a healthier breakfast because of health goals.

3.2 The Anatomy Behind Cognitive Control

Cognitive control is primarily mediated by the prefrontal cortex (PFC), with two major control
systems:
•​ Anterior PFC: Controls abstract, internally driven goals.
•​ Posterior PFC / Lateral PFC: Handles concrete task demands and working memory.
Two important control systems involve the dorsolateral PFC (responsible for active maintenance
and manipulation of information) and the ventrolateral/orbitofrontal PFC (involved in response
inhibition and value-based decisions).

3.3 Cognitive Control Deficits

Patients with frontal lobe damage show a paradox: they appear normal superficially (normal IQ,
perception, movement, speech, long-term memory) but show serious deficits in structured testing
and real-life tasks.

Behavioral Deficits
•​ Perseveration — repeating same response even when wrong
•​ Apathy, distractibility, or impulsivity
•​ Poor decision making and planning
•​ Failure to understand consequences of actions
•​ Difficulty organizing events in memory

📌
•​ Inability to follow rules or social norms
These patients are aware of their problems and can verbally suggest solutions but cannot
organize or act on them — a dissociation between knowing and doing.

Shallice's Real-World Task Study

Patients were asked to perform everyday tasks (buying items, keeping appointments, collecting
information). Results: patients failed to complete tasks — one didn't buy soap because their brand
was unavailable, another left the shopping area unnecessarily, one took a newspaper without
paying. In a 15-minute multi-task, patients focused too long on one activity and neglected others.

Animal Studies (Bianchi)

MCY204C21: Cognitive Neuroscience II | Page 12

Monkeys with frontal lobe damage showed stimulus-driven behavior without purpose: a monkey
would jump to a window (habit) but no longer call others; another would grab a door handle but not
open the door. Their actions became automatic responses to stimuli, lacking goal-directedness.
This was explained as loss of 'focal consciousness,' now understood as a working memory deficit.

Utilization Behavior (Lhermitte)

Patients saw objects and automatically used them without context-appropriate control: a patient saw
a hammer and nail and immediately used them to hang a picture; in an extreme case, a patient
used a needle on the doctor. Patients retained knowledge of object functions but lost the ability to
inhibit context-inappropriate actions.

3.4 Goal-Oriented Behaviour

Types of Actions
•​ Goal-Directed Actions: Based on expected reward and understanding of action-outcome
relationships. The person acts because they know what result it will produce. Examples:
turning on radio to listen to news; studying to get good marks.
•​ Habitual Actions: No longer controlled by reward; become automatic and stimulus-driven.
Triggered by specific situations without thinking. Examples: automatically turning on a broken
car radio; eating junk food when bored.

3.5 Physiological Correlates of Working Memory

Role of the Prefrontal Cortex
Working memory requires (1) accessing stored information and (2) keeping information active for a
short time. The PFC performs both functions. It is crucial for maintaining information across short
delays to guide goal-directed behavior.

Evidence from Monkey Studies — Delayed Response Tasks

Scientists recorded single-neuron activity in the PFC during tasks requiring monkeys to remember
stimulus location or identity across a delay period.
•​ Neurons become active when task begins and continue firing DURING the delay period —
even when the stimulus is gone.
•​ Some neurons start firing only after the delay begins and can remain active for up to 1 minute.
•​ Neurons maintain a mental representation of the stimulus, providing a continuous signal of
what response is needed.

Earl Miller's Working Memory Experiment

Purpose: To determine if PFC neurons code specific information ('What' = object identity, 'Where' =
location) rather than a general memory signal.
•​ Step 1: Monkey fixates on screen.
•​ Step 2: Sample object shown (e.g., a bell) — monkey must remember WHAT it is.
•​ Step 3: First delay — monkey maintains object identity in working memory.

MCY204C21: Cognitive Neuroscience II | Page 13

 •​ Step 4: Two objects appear — monkey identifies the matching one.
•​ Step 5: Location of correct object (left/right) becomes important — now remember WHERE.
•​ Step 6: Second delay — monkey maintains location in working memory.
•​ Step 7: Final response based on location.

Types of PFC Neurons Found

•​ 'What' Cells: Respond to specific objects (identity); remain active during delay.
•​ 'Where' Cells: Respond to location of objects; remain active during spatial delay.
•​ 'What-Where' Cells: Respond to both identity and location; active during first delay for
identity, second delay for location. These integrate both types of information.
Key insight: PFC exhibits task-specific selectivity — neurons respond ONLY to the specific type of
information needed for the current task. Activity depends on using the information, not just
perceiving it. This confirms PFC is NOT a storage unit but a goal-directed processing system.

PFC as Interface (Not Storage)

PFC interacts with temporal cortex (for 'what' information) and parietal cortex (for 'where'
information). It keeps information active through dynamic interaction with these regions during delay
periods. Memory is maintained through PFC + posterior brain areas working together, not stored in
PFC alone.

3.6 Decision Making

Neuroeconomics
An interdisciplinary field combining psychology, economics, and neuroscience to study how the
brain makes decisions. Normative theories describe how we SHOULD decide (optimal outcomes).
Descriptive theories describe how we ACTUALLY decide (often irrational or biased). Our brains
evolved heuristics — fast, automatic, unconscious shortcuts that were useful for survival but may
seem irrational today.

Types of Decisions
•​ Goal-Directed (Action-Outcome): Thinking about what will happen: 'Studying → good
marks.'
•​ Habitual (Stimulus-Response): Automatic behavior: 'Always check phone when it buzzes.'
•​ Model-Free (Habit-based): Direct input → output, no thinking. Example: 'Turn at the big tree
→ reach KJC.'
•​ Model-Based (Planning): Uses a mental map of the world. Example: 'If road is blocked →
choose another route.'

3.7 Value and Decision Making

Before deciding, the brain estimates the value of each option and chooses the one with the highest
value. Value is SUBJECTIVE — it differs between individuals and depends on multiple factors.
•​ Primary Reinforcers: Basic rewards needed for survival: food, water, sex. Innate — the brain
is naturally wired to respond.

MCY204C21: Cognitive Neuroscience II | Page 14

 •​ Secondary Reinforcers: Gain value through association (e.g., money, status, grades).
Valuable because they help obtain primary rewards.

Components of Value
•​ Payoff: How much reward you receive.
•​ Probability: Chance of receiving the reward.
•​ Effort / Cost: How much work is required.
•​ Context: Situational factors (e.g., food more valuable when hungry).
•​ Personal Preference: Individual likes and dislikes.
The brain combines all these factors to compute a final subjective value.

3.8 Representation of Value — Wallis et al. Monkey Study

Researchers recorded brain activity from ACC, LPFC, and OFC as monkeys chose between
options varying in payoff, probability, and effort.
•​ No strict anatomical separation — value is mixed across regions, not stored discretely.
•​ ACC and OFC neurons respond to ALL three factors → compute overall combined value.
•​ LPFC neurons are more specific → respond to only one factor (mostly payoff or probability).
•​ Value signals appear BEFORE motor responses — the brain first decides 'what is valuable,'
then 'what action to take.'

Human fMRI Findings on Value Representation

•​ OFC (Orbitofrontal Cortex): → Payoff/reward value.
•​ Striatum (Basal Ganglia): → Effort encoding.
•​ Lateral PFC: → Probability.
•​ Medial PFC and Parietal Lobe: → Delay (waiting time for reward).
•​ VMPFC / OFC: → Immediate reward (taste value).
•​ DLPFC: → Self-control; resisting temptation for long-term benefit. Strong DLPFC influence =
good self-control.

3.9 Dopamine Activity and Reward Processing

Dopamine neurons in the midbrain (ventral tegmental area and substantia nigra) project to the
striatum, OFC, and PFC. They play a central role in reward learning and prediction.
•​ Reward Prediction Error (Schultz): Dopamine neurons fire for unexpected rewards. As a
stimulus becomes a reliable predictor of reward, dopamine shifts from firing at the reward itself
to firing at the predictive cue. If expected reward does not occur, dopamine activity drops
below baseline — signaling a negative prediction error.
•​ Temporal Difference (TD) Learning: The brain computes the difference between expected
and actual outcomes at each moment. This signal allows learning which actions lead to
rewarding outcomes.
•​ Role in Decision Making: Dopamine modulates the value of actions and stimuli, influencing
which behaviors are reinforced. The striatum (caudate and putamen) uses dopamine signals
to guide habit formation and goal-directed learning.

MCY204C21: Cognitive Neuroscience II | Page 15

Alternative Views of Dopamine Activity
•​ Salience Hypothesis: Dopamine does not merely signal reward — it signals the motivational
salience (importance) of stimuli, whether positive or negative. Unexpected aversive events
can also activate dopamine neurons.
•​ Role in Wanting vs Liking: Berridge proposed that dopamine drives 'wanting' (motivation to
pursue reward) distinct from 'liking' (hedonic pleasure). Dopamine is primarily involved in
motivational drive, not hedonic experience itself.

3.10 Goal Planning — Hierarchical Organisation of the PFC

The PFC is organized hierarchically along a posterior-to-anterior axis. Posterior PFC regions handle
concrete, immediate task demands; anterior PFC (frontopolar cortex) handles abstract, internally
generated goals.
•​ Posterior PFC: Responds to external stimuli; supports immediate behavioral control.
•​ DLPFC (Mid-level): Active maintenance of task-relevant information; task-switching; cognitive
flexibility.
•​ Frontopolar / Anterior PFC: Most abstract level; integrates information across tasks;
maintains multiple goals simultaneously; supports prospective memory and relational
reasoning.

Cognitive Control and Goal Planning

Goal planning requires: developing a plan based on past experience; adapting to changing
environments; monitoring actions; following social rules; inhibiting impulses. The prefrontal-striatal
circuit is critical for both goal planning and monitoring plan execution.

Task-Relevant Information
The PFC selectively maintains task-relevant information and suppresses irrelevant information
during the delay between a cue and a required response. This selective maintenance is what
enables flexible, context-appropriate behavior.

Task Switching
The ability to shift between tasks depends on the DLPFC. Damage causes perseveration —
continuing old rules even when they are no longer appropriate (as demonstrated by the Wisconsin
Card Sorting Test). Task switching requires: inhibiting the old task set; activating the new task set;
updating working memory.

The Medial Frontal Cortex as a Monitoring System

The anterior cingulate cortex (ACC), part of the medial frontal cortex, functions as a performance
monitoring system. It detects conflict between competing responses and signals the need for
increased cognitive control.
•​ The Error-Related Negativity (ERN) is an ERP component generated in the ACC immediately
after errors.
•​ Robert Knight and William Gehring: lateral PFC damage results in patients showing
'corrective' neural activity even for non-errors, suggesting they lost the ability to monitor and
integrate behavior across time.

MCY204C21: Cognitive Neuroscience II | Page 16

•​ The ACC also responds to unexpected outcomes and effort demands, signaling when more
cognitive resources are required.

MCY204C21: Cognitive Neuroscience II | Page 17

UNIT IV: SOCIAL COGNITION
4.1 Anatomical Substrates of Social Cognition
A central debate in social cognition: do we process information about ourselves and others in
separate, overlapping, or the same brain regions? The 'self' is not localized in one area but is a
combination of multiple processes and inputs from inside and outside the body.

Brain Regions for Self-Referential Processing

•​ DLPFC and VMPFC: Thinking about one's own traits, beliefs, past experiences.
•​ Posterior Cingulate Cortex (PCC) and Medial Parietal Cortex: Self-reflection and
autobiographical memory integration.
•​ OFC and ACC and Insula: Processing subjective feelings and emotions as part of the self.
•​ Temporal Lobe: Memory — essential for self-identity.

Brain Regions for Understanding Others

•​ Amygdala: Processes emotions, especially fear and threat; recognizes emotional
expressions; evaluates social relevance.
•​ Superior Temporal Sulcus (STS): Interprets biological motion, gaze direction, and others'
intentions.
•​ Medial PFC (MPFC): Theory of mind — thinking about others' mental states, beliefs,
intentions.
•​ OFC: Evaluates social rewards and punishments; appropriate social behavior.
•​ ACC: Empathy, emotional regulation, detecting social conflicts.
•​ Fusiform Face Area (FFA): Specialized for face recognition and distinguishing individuals.
•​ Mirror Neuron System: Understanding others' actions by internally simulating them;
important for imitation and empathy.
•​ Temporal Poles: Social and emotional knowledge; understanding personal and social
context.
•​ Temporoparietal Junction (TPJ): Perspective-taking; understanding others' beliefs and
intentions (theory of mind).

4.2 Deficits in Social Cognition

Orbitofrontal Cortex (OFC) Damage
Damage to the OFC leads to socially inappropriate behavior and poor social judgment. Such
patients understand social rules verbally but fail to apply them in context.

Autism Spectrum Disorder (ASD)

•​ Definition: A pervasive developmental disorder characterized by impaired social interaction
and communication, with a wide range of severity.

MCY204C21: Cognitive Neuroscience II | Page 18

 •​ Types: Autism, Asperger's syndrome, Childhood disintegrative disorder, Rett syndrome,
PDD-NOS.
•​ Social behavior: Little interest in social interaction; prefers internal thoughts or inanimate
objects.
•​ Routine: Strong need for sameness; distress when routines change.
•​ Repetitive behavior: Self-directed stereotyped actions (flicking, rocking, hand twisting).
•​ Mindblindness (Baron-Cohen, 1995): Difficulty understanding others' mental states —
thoughts, beliefs, and intentions; inability to represent others' perspectives.
•​ Facial perception deficit: Typical children sort faces by emotion; autistic children sort by
physical features (clothing). Reduced attention to eye gaze.
•​ Prosopagnosia link: Face blindness helps explain social perception difficulties in autism.

4.3 Theory of Mind: Understanding the Mental States of Others

Developmental Milestones
•​ 18 months: Joint attention — following another's gaze to a shared object.
•​ 3-4 years: Understanding that others can have false beliefs (Sally-Anne task).
•​ 6-7 years: Understanding second-order beliefs ('She thinks he thinks...').
•​ Mature theory of mind continues developing through adolescence.

Mechanisms for Inferring Others' Thoughts

•​ Simulation Theory / Experience Sharing System (ESS): We understand others by mentally
putting ourselves in their place. Observing behavior creates a similar emotional/physiological
response in us (automatic, unconscious). Linked to the mirror neuron system. Sometimes
conscious: deliberately 'stepping into another's shoes.'
•​ Theory Theory / Mental State Attribution System (MSAS): Used when simulation is not
possible (e.g., person is hiding true feelings). Uses knowledge about the person, their past
behavior, personality, culture, and context to make logical inferences about their mental state.
Both systems work together. The MPFC is active for both self-reference and thinking about others
— we often use our own experiences as a reference to understand others.

MPFC and Self vs Others (Mitchell et al., 2006)

•​ Ventral MPFC: active when thinking about SELF and SIMILAR others.
•​ Dorsal MPFC: active when thinking about DISSIMILAR others.
•​ Ochsner et al. (2005): Same MPFC region active for self and romantic partner — due to deep,
emotionally rich knowledge of both.

Empathy and Its Neural Basis

Empathy = ability to understand and respond to others' experiences. The Perception-Action Model
(Preston & de Waal): observing another's emotional state automatically activates the same state in
the observer, producing corresponding bodily and emotional responses.
•​ Mirror neurons: Enable internal simulation of others' states (embodied simulation). Connect
with emotional processing areas for empathy.

MCY204C21: Cognitive Neuroscience II | Page 19

 •​ Insula: Crucial link between the mirror neuron system and the limbic system. Active both
when experiencing AND observing disgust (Singer study: active for self pain and observed
pain in loved one).
•​ ACC: Active for both personal pain and observed pain. Same neurons respond in both cases
(single-unit recordings).
•​ Somatosensory Cortex: Mirrors painful and non-painful touch; damage impairs
understanding others' emotions.

4.4 Neural Correlates of Mental State Attribution

Key Brain Regions
•​ MPFC: Inferring hidden mental states; forming impressions. Mitchell et al. (2004): MPFC
highly active during impression formation (inferring personality from behavior) compared to
sequencing task.
•​ rTPJ (Right Temporoparietal Junction): Discovered by Rebecca Saxe (MIT). Specifically
involved in theory of mind — understanding others' beliefs and perspectives. Identified using
the Sally-Anne false-belief paradigm: 'Where will Sally look for her object?' — requires
understanding false beliefs. rTPJ activates for false-belief stories but NOT for false-information
stories without mental state reasoning.
•​ STS: Processes head position and eye gaze direction. Responds to the MEANING of gaze
(toward vs. away from object), not just eye movement. Greater activation for direct eye contact
than averted gaze.
•​ Temporal Poles: Store social and emotional knowledge for attribution.

Importance of Eye Gaze

Humans uniquely follow eye gaze rather than head direction, enabled by prominent white sclera.
Eye gaze is a critical social cue for understanding attention, intentions, and mental states.
Individuals with autism show reduced attention to eye gaze (Adolphs & Spezio studies).

Autism and Mental State Attribution

Autistic individuals show reduced rTPJ and MPFC activation during theory of mind tasks. They
struggle with the Sally-Anne task (often answering based on WHERE the object actually is, not
where Sally BELIEVES it is). They rely less on eye gaze and facial expressions to infer mental
states.

4.5 Social Knowledge: Representations and Making Decisions

Representation of Social Knowledge
Social behavior does not follow fixed, universal rules. The same behavior is appropriate in some
contexts and inappropriate in others. The frontal lobes, particularly the OFC, are crucial for
evaluating situations and applying appropriate social rules.

Stone et al. — Social Faux Pas Task

MCY204C21: Cognitive Neuroscience II | Page 20

Patients with OFC damage were asked to evaluate a scenario (Jeannette/Anne story) where a
character unknowingly says something inappropriate. Results: OFC patients performed significantly
WORSE — they understood basic emotional reactions (Jeannette feels bad about breaking bowl)
but FAILED to grasp social intention (Anne trying to reassure her, not hurt her). They could not
integrate context and intention in social reasoning.

Beer (Jennifer Beer) Studies

•​ Study 1: OFC patients in structured conversation with strangers — violated social norms even
though they knew those norms beforehand. Clear dissociation between knowledge and
application.
•​ Study 2 (Teasing task): OFC patients created unflattering nicknames and used them in
inappropriately intimate tones — and felt PROUD, not embarrassed. This lack of
embarrassment prevented them from using emotional feedback to adjust behavior.
The OFC integrates social knowledge, contextual cues, and emotional feedback to produce
appropriate social behavior. Without OFC, patients fail to apply social rules, cannot recognize their
own mistakes, and cannot adjust future behavior (due to absent self-insight and lack of emotional
feedback).

VMPFC and Social Decision Making

•​ Fellows & Farah (2007): VMPFC damage impairs value-based decisions even for simple
preferences (colors, foods, faces) — broader than just uncertainty.
•​ Reversal Learning: OFC damage prevents adapting behavior when value of stimulus
changes (e.g., when previously rewarded becomes punishing). Cannot use negative feedback
to adjust social behavior.
•​ Schoenbaum et al.: OFC signals when expected outcomes are violated, interacting with
amygdala to update behavior. Wheeler & Fellows: Damage to ventromedial frontal selectively
impairs learning from NEGATIVE feedback (learning from positive feedback intact).
•​ FTLD Patients: Frontotemporal lobar degeneration → socially inappropriate behavior; rate
negatively biased social scenarios as more acceptable than controls. Healthy individuals show
increased VMPFC activation for negative social scenarios.

4.6 Neuroeconomics
Integrates psychology, neuroscience, and economics to explain value-based decisions. Traditional
economics assumes rational behavior; neuroeconomics recognizes that emotions and social factors
shape decisions.
•​ Framing Effect: When given a choice between guaranteed gain vs. gamble → people avoid
risk. With guaranteed loss vs. gamble → people take risks, even with equivalent monetary
outcomes. This occurs because guaranteed loss generates strong negative emotional
response (amygdala activation for emotion-driven decisions; OFC for rational evaluation — De
Martino et al., 2006).
•​ Ultimatum Game: One player proposes money split; other accepts or rejects. People reject
unfair offers even at personal financial cost. Unfair offers activate the INSULA (negative
emotions); increased insula activity predicts rejection. Functions to protect social reputation
and discourage unfair behavior.

MCY204C21: Cognitive Neuroscience II | Page 21

 •​ Prisoner's Dilemma: Cooperation → activates nucleus accumbens, OFC, ACC, and caudate
nucleus (reward-related regions). Prosocial behavior is intrinsically rewarding.
•​ Nucleus Accumbens / OFC: Encode subjective value for primary rewards (food), secondary
rewards (money), AND social rewards (observing others succeed, social agreement).
Prosocial outcomes are inherently rewarding.

4.7 Moral Decisions

Greene et al. (2001, 2004) used the trolley problem to study moral decision making. The key
variable is PERSONAL ENGAGEMENT — the degree to which a person feels directly responsible.
•​ Trolley Problem: Divert runaway trolley: kill 1 to save 5 by pulling a switch (impersonal). Most
people find this acceptable.
•​ Footbridge Dilemma: Push a person off a bridge to stop the trolley: kill 1 to save 5
(personal). Most find this morally unacceptable, despite identical numerical outcome.
•​ Impersonal Moral Decisions: Activate lateral prefrontal cortex and parietal lobes (cognitive
processes — working memory, reasoning).
•​ Personal Moral Decisions: Activate medial frontal cortex, posterior cingulate gyrus, and
amygdala (emotional processing and social cognition).
Moral judgment is not purely rational but involves a dynamic interplay between cognitive control
systems and emotional processing. When emotional engagement is high, people reject actions that
feel personally harmful even if they lead to better overall outcomes.
•​ Utilitarian vs. Deontological: Utilitarian: outcome-based — cutting the rope was justified
(one survives). Deontological: rule-based — cutting the rope violates moral duty. Both
perspectives are reflected in real neural activity.

MCY204C21: Cognitive Neuroscience II | Page 22

UNIT V: CONSCIOUSNESS AND FREE WILL
5.1 Defining Consciousness
Consciousness is one of the most challenging topics in neuroscience and philosophy. The
International Dictionary of Psychology (Sutherland) defines it as 'the having of perceptions,
thoughts, and feelings; awareness.' Consciousness is fascinating but elusive — difficult to specify
what it is, what it does, or why it evolved.

Historical Perspectives
•​ Dualism (Descartes): Mind and body are two distinct phenomena. Conscious experience is
nonphysical and beyond the scope of science.
•​ Materialism: Both mind and body are physical. By understanding brain function, the mind can
be understood.
Steven Pinker (1997) identified three dimensions of consciousness: Self-knowledge (awareness of
one's own mental states), Access to information (ability to report content of mental experience), and
Sentience (subjective experience, qualia, raw feelings — the 'hard problem').

Damasio's Categories (Antonio Damasio)

•​ Core Consciousness: Basic awareness — the organism is alive, awake, alert, aware of one
moment ('now') and one place ('here'). Not concerned with past or future. This is the
consciousness 'ON' switch.
•​ Extended Consciousness: Provides an elaborate sense of self; places the self in historical
time; includes thoughts of past and future; depends on the gradual build-up of an
autobiographical self from memories and expected future experiences.
Consciousness has nested layers of organizational complexity (Damasio & Meyer, 2008).
Wakefulness is necessary for consciousness but consciousness is not necessary for wakefulness.

5.2 Anatomical Substrates of Consciousness

Brainstem
•​ Primary role: homeostatic regulation (medulla oblongata + pons).
•​ The consciousness 'ON' switch — modulates wakefulness.
•​ Disconnection from higher structures → loss of consciousness.

Pons and Reticular Activating System (RAS)

•​ Reticular Formation: heterogeneous nuclei involved in motor control, cardiovascular control,
pain modulation, and filtering irrelevant stimuli.
•​ RAS has two pathways: Dorsal (via intralaminar nucleus of thalamus to cortex) and Ventral
(via hypothalamus and basal forebrain to cortex).
•​ Involved in arousal, sleep-wake cycle regulation, and mediating attention.

MCY204C21: Cognitive Neuroscience II | Page 23

 •​ Locus Coeruleus: Main site of norepinephrine production in pons. Prevents sleep by
activating the cortex when active.

Thalamus
•​ Intralaminar Nuclei (ILN) of the thalamus play a key role in CORE consciousness.
•​ Small bilateral lesions to the ILN permanently turn off core consciousness; a lesion to one
alone does not.
•​ If neurons connecting the thalamic ILN and brainstem are severed, core consciousness is lost.
•​ All sensory input about body and world passes through the thalamus.
•​ Thalamocortical loops (thalamus ↔ cortex) coordinate activity throughout the cortex,
contributing to consciousness.
•​ Lesions anywhere from brainstem up to cortex can disrupt core consciousness.

Cerebral Cortex
•​ In concert with brainstem and thalamus, maintains wakefulness and contributes to selective
attention.
•​ Extended consciousness begins with cortical contributions that help generate the core of self
(records from memory of past activities, emotions, experiences).
•​ Damage to cortex → loss of a specific ability (not consciousness itself) — e.g., damage to
visual cortex → blindness; parietal damage → neglect.
•​ Patients with unilateral parietal lesions: neglect of half of space (not conscious of half the
world around them).

5.3 Conscious vs. Unconscious Processing

The vast majority of mental processes that control and contribute to our conscious experience
happen OUTSIDE conscious awareness. We are conscious only of the CONTENT of mental life,
not of what GENERATES that content.
•​ Conscious processing: Conjunction — putting together awareness of the stimulus with its
identity, location, orientation, or other features.
•​ Unconscious processing: Disjunction — separating awareness from features, such that
even when unaware of a stimulus, people can still respond to its features at above-chance
levels.

Blindsight (Larry Weiskrantz, Oxford, 1974)

Patients with lesions in the visual cortex (striate cortex / V1) can respond to visual stimuli presented
in the blind part of their visual field. They DENY being able to see anything, yet perform above
chance on forced-choice tasks about those stimuli.
•​ These patients have ACCESS to visual information but do not EXPERIENCE it (no sentience).
•​ Ned Block used blindsight to distinguish sentience (subjective experience) from access (ability
to report on mental content).
•​ Weiskrantz proposed subcortical and parallel pathways (e.g., superior colliculus) are
responsible.

MCY204C21: Cognitive Neuroscience II | Page 24

 •​ Wessinger & Fendrich at Dartmouth: evidence that the damaged primary visual pathway is

📌
INVOLVED — not only secondary visual systems.
Blindsight supports the view that perception can happen in the absence of sensation.
Perceptual decisions routinely result from processes outside conscious awareness.

Unilateral Neglect Studies

Patients with right parieto-occipital damage show unilateral neglect — they cannot name stimuli in
the left visual field. However, when asked to judge whether two stimuli (one in each field) are 'same
or different,' they can do so correctly — yet still deny having seen the left field stimulus.
Volpe et al. (1979): In double simultaneous stimulation (DSS), patients report only the right field
stimulus verbally, yet can perform inter-field comparisons correctly — demonstrating that
unconscious processing of the extinguished stimulus occurs.

5.4 The Extent of Subconscious Processing

Complex, high-level information can be processed outside conscious awareness.
•​ Neglect Category Priming Study: A fruit/animal picture briefly presented to the right visual
field. Then a related picture in the left (neglected) field. Patients denied seeing the left field
picture, yet responded FASTER when pictures were related — showing high-level semantic
information was exchanged between processing systems outside conscious awareness.
•​ Nisbett & Ross (1980) Word-Pair Study: Healthy participants exposed to 'ocean-moon' word
pairs. Later said 'Tide' to 'detergent' but didn't know why — gave false explanations. Shows
unconscious priming of healthy brains.
•​ Marcel's Masking Paradigm: Brief presentation of a word followed immediately by a masking
stimulus (crosshatch). Participants could NOT detect if a word had been shown (chance
performance) but showed LEXICAL PRIMING — related words produced faster response
times. Shows semantic processing occurs without conscious awareness.
•​ Tower of Hanoi Studies: Problem solving operates on TWO levels simultaneously: conscious
(recalling rules, facts) and unconscious (how those facts were encoded in memory). We are
aware only of the conscious level.

General Principle
Many perceptual and cognitive activities occur outside conscious awareness. Unconscious
processes are not limited to subcortical/secondary systems — they likely happen at the cortical
level. We can access information we are not consciously aware of (access without sentience).

5.5 Gaining Access to Consciousness

Pinker (1997) argued that evolutionary pressures gave rise to access-consciousness by imposing
constraints: cost of space, time, and metabolic resources (glucose). Only information relevant to the
problem at hand should enter consciousness — this appears to be how the brain is organized.

Four Features of Access-Consciousness (Pinker)

•​ Brimming with sensations (sensory content — colors, smells, pain).

MCY204C21: Cognitive Neuroscience II | Page 25

 •​ Ability to move information in and out of awareness using an attentional spotlight.
•​ Information always comes with salience (emotional coloring).
•​ The 'I' — a self that decides what to do with incoming information.

5.6 Sentience
Damasio (2011): Consciousness is a mind state in which the regular flow of mental images has
been enriched by subjectivity — mental images that represent body states.
•​ The body continuously signals the brain and is signaled back in a perpetual resonant loop.
•​ Mental images about the self (body) are connected to the body, and thus are 'felt.'
•​ Because these images are felt, the organism senses that its thoughts are its own —
formulated in the perspective of the organism.

📌
•​ Sentience does NOT imply meta self-awareness (being aware that one is aware of oneself).
The 'hard problem of consciousness' (Chalmers): explaining why any physical process gives
rise to subjective experience (qualia) at all. Some philosophers believe this will never be fully
explained.

5.7 Neurons, Neuronal Groups, and Conscious Experience

Newsome's Studies (Area MT Motion Detection)
William Newsome at Stanford studied how neural events in area MT (motion detection) correlate
with perceptual decisions in monkeys. Key findings:
•​ A single neuron in area MT was as sensitive to motion changes as the whole monkey's
psychophysical performance.
•​ Microstimulation of area MT neurons tilted the monkey toward making the correct decision on
a perceptual task — an artificial signal inserted into the nervous system influenced thinking.
•​ This does not mean a single neuron IS the decision — they tapped into part of a neural loop.
Stimulation at different points in the loop creates different subjective experiences.

Libet's Backward Referral Hypothesis

Benjamin Libet investigated neural time factors in conscious and unconscious processing. He found
that brain activity preceding a voluntary movement (readiness potential) begins up to 500ms
BEFORE the person is conscious of deciding to move.
•​ Haynes et al. (Soon et al., 2008): fMRI shows decisions can be encoded in brain activity up to
10 seconds before they enter awareness.
•​ Libet's conclusion: enough time elapses between the awareness of the intent to act and the
actual beginning of the act that we can OVERRIDE inappropriately triggered behavior. This
ability to detect and correct errors is the basis for FREE WILL.

Error Detection and Correction

•​ Robert Knight and William Gehring: characterized role of frontal lobe in checking and
correcting errors using ERPs.

MCY204C21: Cognitive Neuroscience II | Page 26

 •​ In letter discrimination tasks, healthy volunteers showed expected corrective neural activity in
the anterior cingulate cortex (ACC).
•​ Patients with lateral PFC damage showed corrective activity for BOTH errors AND non-errors
— losing the ability to monitor and integrate behavior across time.

5.8 The Emergence of the Brain Interpreter

Gazzaniga's split-brain research revealed that the left hemisphere generates explanations and
narratives for actions, creating an 'interpreter' — a system that creates plausible (though sometimes
inaccurate) stories for behaviors initiated by the right hemisphere or subcortical systems. This
interpreter may be fundamental to the experience of a unified, coherent consciousness.
•​ Left-Hemisphere Consciousness: The left hemisphere is the major source of the verbal,
narrative self. It monitors the world and attempts to make sense of events by creating
cause-and-effect explanations.
•​ Right-Hemisphere Consciousness: The right hemisphere processes information more
holistically, is better at spatial tasks and implicit associations, and plays a greater role in
emotional processing and insight.

5.9 Free Will

Determinism and Physics
Determinism: the philosophical belief that all current and future events, including human cognition,
decisions, and behavior, are caused by preceding events combined with the laws of nature. If so, all
behavior is predictable in principle, and individuals are not personally responsible for their actions.
Newton's laws suggested the universe is a clockwork mechanism — fully determined.

Chaos Theory
Henri Poincaré (1889) showed that even tiny differences in initial conditions produce wildly different
outcomes in systems of three or more interacting bodies — the 'n-body problem.' Chaos theory
(rediscovered by Edward Lorenz) demonstrates that deterministic systems can be practically
unpredictable.
•​ Lorenz: minute variations in input data (rounding 0.506127 to 0.506) produced wildly different
computer model outputs.
•​ The 'butterfly effect' (1972 lecture: 'Does the flap of a butterfly's wings in Brazil set off a
tornado in Texas?') — small uncertainties overwhelm long-term predictions.
•​ A deterministic process can be practically UNPREDICTABLE. Chaotic behavior observed in
electrical circuits, population growth, and action potentials in neurons.

Quantum Theory
Quantum mechanics showed that atoms do not obey Newton's classical laws. The Schrödinger
equation can only predict PROBABILITIES, not exact positions/momenta of particles simultaneously
(Heisenberg Uncertainty Principle). This fundamentally undermines determinism at the subatomic
level.

MCY204C21: Cognitive Neuroscience II | Page 27

 •​ At the macroscopic level, Newton's laws emerge from quantum reality and appear
deterministic.
•​ Physics shifted from an inherently deterministic to an inherently nondeterministic worldview at
the fundamental level.

Emergence
A complex system, composed of many interconnected parts that self-organize, exhibits properties
NOT present in or predictable from its individual components. Examples: ant colonies, brain,
climate, human societies.
•​ Weak Emergence: New properties arise from elemental interactions and ARE reducible to
components. One level can be predicted from another (e.g., Newton's laws could theoretically
be derived from quantum mechanics).
•​ Strong Emergence: New property is irreducible — more than the sum of its parts. The laws
of one level CANNOT be predicted from another level. The whole is truly more than the sum of
its parts.
Newton's laws are themselves EMERGENT — when quantum matter aggregates into macroscopic
objects, a new level of organization emerges with its own laws. Consciousness may be an
emergent property of brain activity that cannot be fully explained by reducing it to neural firing.
📌 Nobel physicist P.W. Anderson (1972), 'More Is Different': 'The ability to reduce everything to
simple fundamental laws does not imply the ability to start from those laws and reconstruct the
universe.'

Multiple Realizability
Eve Marder's work with spiny lobsters: she mapped the entire neural network of the lobster's gut
motility system. After 20 million stimulations of possible network combinations, she found that
100,000-200,000 different neural tunings could produce the exact same gut behavior.
•​ Multiple realizability: many different ways to implement a system to produce one behavior.
•​ Implication for neuroscience: even a simple nervous system cannot be fully understood
through reductionist analysis alone — analyzing individual neurons and synapses will not tell
us HOW the system actually works.
•​ For the human brain, the number of possible neural tunings for a single behavior would be
astronomically large.
This challenges the reductionist assumption that macro behavior can always be derived from micro
analysis.

Mental States and Brain Processing

Vohs & Schooler experiment: students primed with DETERMINISTIC statements ('We are biological
computers') were more likely to cheat on a test than those primed with FREE WILL statements.
Mental state (belief about free will) affected subsequent behavior.
•​ Baumeister et al. (2009): reading deterministic passages resulted in more aggressive and less
helpful behavior toward others.
•​ Belief in free will motivates people to control automatic impulses and exercise self-control.
•​ Most neuroscientists resist top-down causation (the idea that a mental state, as an emergent
property, can constrain the physical brain that produced it).

MCY204C21: Cognitive Neuroscience II | Page 28

The classic puzzle: Physical state P1 → Mental state M1 → Physical state P2 → Mental state M2.
Does M1 somehow constrain P2? If it doesn't, our mental life is doing no work. The challenge is
explaining how emergent mental states might exert downward causation.

Symmetry Breaking
Mental states are not chosen freely in the libertarian sense — they are deterministic, modularized,
and driven by hundreds or millions of interacting physical processes. But the system is
unpredictable because small fluctuations acting on the system at critical points (symmetry breaking)
determine which of several equally likely outcomes will occur. Our 'interpreter' then claims we freely
made a choice.
•​ The key insight: neither the mental layer nor the neuronal layer functions without the
participation of the other.
•​ Freedom is not the absence of causation but the interaction between hierarchical layers of
brain organization.

— END OF NOTES —
MCY204C21: Cognitive Neuroscience II | All 5 Units

MCY204C21: Cognitive Neuroscience II | Page 29