UNIT 3

PHYSIOLOGICAL BASIS OF MEMORY

Relational Learning
From: Foundations of Physiological Psychology (6th Edition)
Chapter 12: Learning and Memory

TOPICS COVERED:
1. Human Anterograde Amnesia – Basic Description
2. Spared Learning Abilities in Anterograde Amnesia
3. Declarative and Nondeclarative Memories
4. Anatomy of Anterograde Amnesia
5. Anterograde Amnesia as Failure of Relational Learning
6. Role of the Hippocampal Formation in Spatial Memory
7. Relational Learning in Laboratory Animals
1. INTRODUCTION TO RELATIONAL LEARNING

Definition: Relational learning refers to the establishment and retrieval of memories that link
together multiple stimuli, events, and episodes in time and place. Unlike simple forms of learning
(perceptual, stimulus-response, motor), relational learning involves complex associations between
memories.

Key Point: Most real-world memories are relational in nature. Seeing a photograph of an old friend
can trigger associated memories of conversations, shared experiences, the friend's name, and how
to pronounce it — demonstrating the complex, interconnected nature of relational memory.

Why Relational Learning is Significant

• Simple forms of learning (perceptual, stimulus-response, motor) involve strengthened
connections between specific neural circuits.
• In relational learning, neural circuits in the visual association cortex are connected to
circuits in many other brain regions.
• Relational learning includes the establishment and retrieval of episodic memories —
memories of specific events and their contexts.
• Understanding relational learning helps explain the nature of anterograde amnesia, one of
the most dramatic and intriguing consequences of brain damage.
2. HUMAN ANTEROGRADE AMNESIA

2.1 Definition and Overview

TERM DEFINITION
Anterograde Amnesia Amnesia for events that occur AFTER some disturbance to the
brain, such as head injury or certain degenerative brain diseases.
At first glance it appears to be the inability to learn new information.
Retrograde Amnesia Amnesia for events that PRECEDED some disturbance to the brain,
such as a head injury or electroconvulsive shock.
Pure Anterograde A rare form where only post-damage events are forgotten; usually
Amnesia accompanied by some retrograde amnesia too.
Korsakoff's Syndrome Permanent anterograde amnesia caused by brain damage resulting
from chronic alcoholism or malnutrition.

Critical Insight: Anterograde amnesia is NOT simply the inability to learn. The basic abilities of
perceptual learning, stimulus-response learning, and motor learning remain INTACT. What is lost is
complex relational learning.

2.2 Historical Background: Korsakoff's Syndrome

In 1889, Russian physician Sergei Korsakoff first described a severe memory impairment caused
by brain damage. Key characteristics of this syndrome include:
• Severe anterograde amnesia: Patients appear unable to form new memories.
• Intact old memories: They can still remember events that happened long before their brain
damage.
• Normal conversation: They can converse normally and engage with the present moment.
• Loss of post-damage memories: They cannot remember events that happened after the
damage.
• Common cause: Usually (but not always) a result of chronic alcoholism, which leads to
thiamine deficiency and subsequent brain damage.

CASE STUDY: Patient H.M. — The Most Famous Case in Neuroscience

Background: H.M. (Henry Molaison, 1926-2008) suffered from extremely severe epilepsy
that could not be controlled by high doses of anticonvulsant medication. His epilepsy was
caused by a head injury he received when struck by a bicycle at age nine.
The Surgery (1957): Neurosurgeon William Beecher Scoville performed bilateral removal of
 the medial temporal lobe in an attempt to treat H.M.'s epilepsy. The surgery successfully
treated his seizure disorder but produced a serious and permanent memory impairment.
Discovery: Scoville and Brenda Milner (1957) reported that bilateral removal of the medial
temporal lobe produced a memory impairment identical to Korsakoff's syndrome. Because
H.M. was reasonably intelligent and not psychotic, his postoperative deficit was discovered
immediately.
Consequence for Surgery: Once it was discovered that bilateral medial temporal
lobectomy causes anterograde amnesia, neurosurgeons stopped performing them and are
now careful to operate on only one temporal lobe.
Research Value: Because of his relatively pure amnesia, H.M. was extensively studied by
Milner and her colleagues, who based foundational conclusions about memory on his
pattern of deficits.

2.3 Three Foundational Conclusions from H.M.'s Case

Milner (1970) and colleagues drew three critical conclusions based on H.M.'s pattern of memory
deficits. These conclusions, though later refined by further research, remain foundational:

1. Conclusion 1: The hippocampus is NOT the location of long-term memories, nor is it

necessary for retrieval of long-term memories.
• Evidence: H.M. could remember events from early in his life, knew how to talk, how to dress
himself, and retained pre-surgical knowledge.
• If the hippocampus stored LTM, all of this would have been lost.

2. Conclusion 2: The hippocampus is NOT the location of immediate (short-term)

memories.
• Evidence: H.M. could carry on conversations — he could remember what the other person
said long enough to think of and give a reply.
• If hippocampus were needed for short-term memory, even momentary recall would have
been impossible.

3. Conclusion 3: The hippocampus IS involved in converting short-term memories into

long-term memories.
• Evidence: H.M. could understand and hold information in mind as long as he thought about
it, but no permanent record of this information was ever made.
• This process — converting STM to LTM — is called CONSOLIDATION.

2.4 Short-Term Memory vs. Long-Term Memory

Many psychologists believe learning consists of at least two stages:
 Feature Short-Term Memory (STM) Long-Term Memory (LTM)
Capacity Limited — only a small amount Unlimited (or enormously
of information large)
Duration Temporary — maintained by Relatively permanent
rehearsal
Mechanism Neural activity (ongoing) Synaptic changes (structural)
Loss Information lost once Information retained without
rehearsal stops rehearsal
H.M. INTACT — could hold IMPAIRED — no new LTMs
information momentarily formed

Consolidation: The process by which short-term memories are converted into long-term memories
— essentially 'making memories solid.' The hippocampal formation's role in memory is this
consolidation process.
3. SPARED LEARNING ABILITIES IN ANTEROGRADE AMNESIA

H.M.'s memory deficit, while striking, does NOT represent a total failure of learning. Spiers,
Maguire, and Burgess (2001) reviewed 147 cases of anterograde amnesia and found that patients
with anterograde amnesia are capable of THREE of the four major types of learning:

• (1) Perceptual Learning — recognizing previously encountered stimuli

• (2) Stimulus-Response Learning — classical and instrumental conditioning
• (3) Motor Learning — acquiring new motor skills and sequences

What is lost: The fourth type — RELATIONAL (declarative) learning — the ability to consciously
recollect the circumstances of learning.

3.1 Spared Perceptual Learning

Broken Drawings Task (Milner, 1970)
• Subjects were shown incomplete drawings of objects, presented in sets from least complete
(Set I) to most complete (Set V).
• If a subject could not recognize the figure in the least complete set, they were shown
progressively more complete sets until they identified it.
• One hour later, subjects were retested, starting with Set I.
• H.M. showed CONSIDERABLE IMPROVEMENT when retested — demonstrating
perceptual learning.
• Four months later, he STILL showed improvement, demonstrating long-term retention of
perceptual learning.
• Crucial finding: Although he showed clear learning, H.M. DENIED ever having seen the
drawings before.

Face and Melody Recognition (Johnson, Kim & Risse, 1985)

• Korean songs were played to amnesic patients.
• When tested later, patients PREFERRED those melodies over ones they had not heard
before.
• Researchers also showed photographs of two men with stories — one described as nice,
one as nasty.
• Twenty days later, amnesic patients said they LIKED the picture of the 'nice' man better than
the 'nasty' one.
• Critical finding: The patients showed clear perceptual preference learning, but they did
NOT recognize that they had heard the melodies or seen the photographs before.
3.2 Spared Stimulus-Response Learning
Classical Conditioning (Woodruff-Pak, 1993)
• H.M. and another amnesic patient could acquire a classically conditioned eyeblink response.
• H.M. even showed RETENTION of the task TWO YEARS LATER.
• He acquired the response in ONE-TENTH the number of trials needed previously —
demonstrating robust retention of the conditioned response.

Instrumental Conditioning (Sidman, Stoddard & Mohr, 1968)

• H.M. was successfully trained on a visual discrimination task.
• He was given pennies for correct responses in an instrumental conditioning task.
• Once learned, when experimenters interrupted him, had him count pennies, then asked
what he was supposed to do — he had no idea.
• Remarkable: When stimuli were turned on again, he IMMEDIATELY made the correct
response — demonstrating intact stimulus-response memory without declarative memory of
having learned it.

3.3 Spared Motor Learning

Sequence Learning Task (Reber & Squire, 1998)
• Subjects with anterograde amnesia sat in front of a computer screen.
• An asterisk appeared — apparently randomly — in one of four locations.
• Their task: press the button corresponding to the location of the asterisk.
• As soon as they pressed, the asterisk moved to a new location.
• The sequence of button presses (e.g., DBCACBDCBA) was NOT random — it was a ten-
item repeated sequence.
• With practice, subjects became FASTER AND FASTER at the task.
• When the sequence was changed, their performance DECREASED — proving they had
learned the specific sequence.
• Critical finding: Amnesic subjects learned this task JUST AS WELL as normal subjects, but
were COMPLETELY UNAWARE there was a sequence — they thought the asterisk moved
randomly.

CASE STUDY: Summary: What H.M. Could and Could Not Do

Type of Learning Task Example Performance Awareness?
Perceptual Broken drawings NORMAL No — denied seeing
recognition before
Perceptual Face/melody NORMAL No — no recognition
preference
Stimulus-Response Eyeblink conditioning NORMAL No — didn't
remember training
Motor Button-press NORMAL No — thought random
sequence
Declarative/Relational Recalling new SEVERELY N/A
episodic events IMPAIRED
4. DECLARATIVE AND NONDECLARATIVE MEMORIES

The distinction between what amnesic patients CAN and CANNOT learn reveals the basic
organization of the learning process.

4.1 Definitions

TERM DEFINITION
Declarative Memory Memory that can be verbally expressed; memory for events in a
person's past. Explicitly available to conscious recollection as facts,
events, or specific stimuli. The term 'declarative' comes from
'declare' — to proclaim or announce.
Nondeclarative Memory Memory whose formation does NOT depend on the hippocampal
formation; a collective term for perceptual, stimulus-response, and
motor memory. These memories appear to operate automatically.
Explicit Memory Another term for declarative memory — conscious, intentional
recollection of information.
Implicit Memory Another term for nondeclarative memory — memory expressed
through performance, without conscious awareness.
Episodic Memory The most complex form of declarative memory — memories of
particular episodes consisting of collections of perceptions of events
organized in time, identified by a particular context.

4.2 Key Characteristics of Declarative Memory

• Explicitly available to conscious recollection
• Memories of FACTS, EVENTS, or SPECIFIC STIMULI
• Requires the HIPPOCAMPAL FORMATION for encoding
• Patients with anterograde amnesia CANNOT form new declarative memories
• Includes EPISODIC memory (specific autobiographical events) and SEMANTIC memory
(general knowledge about the world)
• Example: Remembering that you attended a lecture this morning, what was discussed, who
sat next to you

4.3 Key Characteristics of Nondeclarative Memory

• Does NOT require conscious recollection or intentional memorization
 • Does NOT depend on the hippocampal formation
• Appears to OPERATE AUTOMATICALLY
• Does not seem to include FACTS or EXPERIENCES — instead, it CONTROLS
BEHAVIORS
• PRESERVED in patients with anterograde amnesia
• Example: Knowing how to ride a bicycle — the motor skills are intact even if the memory of
learning to ride is gone

4.4 A Practical Illustration: Learning to Ride a Bicycle

This example beautifully illustrates the distinction:
• Nondeclarative (motor) memory: We learn to ride — the automatic adjustments of hands
and body that keep our center of gravity above the wheels. This is preserved even in
amnesia.
• Declarative memory: We consciously develop declarative memories about our attempts —
who helped us learn, where we rode, how we felt, how many times we fell. This is lost in
anterograde amnesia.

4.5 Experimental Evidence: Graf, Squire & Mandler (1984)

This classic experiment directly demonstrated the explicit/implicit memory distinction in amnesic
patients:

Procedure
4. Showed amnesic and nonamnesic subjects lists of six-letter words to study carefully.
5. Subjects were asked to rate how much they liked each word (to ensure they thought about
each word).
6. Administered two types of memory tests:
• Explicit memory test: Subjects were asked to RECALL the words they had seen
(declarative memory condition).
• Implicit memory test: Subjects were shown cards with the first THREE LETTERS of the
words (e.g., DEF___) and asked to say the first word that came to mind — not to recall the
words from the study list.

Results
• Amnesic subjects remembered FEWER THAN HALF as many words as control subjects on
the EXPLICIT (declarative) recall test.
• Both groups performed EQUALLY WELL on the IMPLICIT (nondeclarative) word-completion
test.
Conclusion
This demonstrates a clear double dissociation: anterograde amnesia selectively impairs explicit
(declarative) memory while leaving implicit (nondeclarative) memory intact.

4.6 Table of Declarative and Nondeclarative Memory Tasks

Memory Type Task Type of Learning Preserved in
Amnesia?
Declarative Remembering past Episodic/Semantic NO
experiences
Declarative Learning new words Semantic NO
(explicitly)
Declarative Recalling word from Semantic NO
DEF___
Nondeclarative Broken drawings Perceptual YES
recognition
Nondeclarative Recognizing faces Perceptual YES
Nondeclarative Recognizing melodies Perceptual YES
Nondeclarative Eyeblink conditioning Stimulus-Response YES
Nondeclarative Choose circle task Stimulus-Response YES
Nondeclarative Sequence of button Motor YES
presses
Nondeclarative Word completion Stimulus-Response YES
(DEF___)
5. ANATOMY OF ANTEROGRADE AMNESIA

To understand anterograde amnesia, we must examine the brain damage that produces it. One fact
is clear: damage to the hippocampus, or to regions that supply its inputs and receive its outputs,
causes anterograde amnesia.

5.1 The Hippocampal Formation: Anatomy

Key Input Pathway
• The most important input to the hippocampal formation comes from the ENTORHINAL
CORTEX.
• The entorhinal cortex receives inputs from the CINGULATE CORTEX and ALL REGIONS
OF THE ASSOCIATION CORTEX, either directly or via two adjacent regions of limbic
cortex:
• Perirhinal cortex: A region of limbic cortex adjacent to the hippocampal formation that,
along with the parahippocampal cortex, relays information between the entorhinal cortex
and other regions of the brain.
• Parahippocampal cortex: A region of limbic cortex adjacent to the hippocampal formation
that, along with the perirhinal cortex, relays information between the entorhinal cortex and
other regions of the brain.
• The hippocampal formation also receives information from the AMYGDALA — this may be
responsible for the role that emotions play in memories.

Key Output Pathway

• Outputs of the hippocampal system are relayed BACK through the entorhinal, perirhinal, and
parahippocampal cortex to the SAME REGIONS that provide inputs — the cingulate cortex
and all regions of the association cortex.
• The hippocampal formation also receives input from subcortical regions via the FORNIX,
carrying dopaminergic axons from the ventral tegmental area, noradrenergic axons from the
locus coeruleus, serotonergic axons from the raphe nuclei, and acetylcholinergic axons from
the medial septum.
Analogy: The cortical inputs select and provide specific information; the subcortical inputs (via
fornix) modulate the functions of the hippocampal formation — like an antenna supplying
information vs. switches controlling a radio's functions.

5.2 Critical Evidence: The Role of Field CA1

Patient R.B.: Evidence from Anoxia (Zola-Morgan, Squire & Amaral, 1986)
 CASE STUDY: Patient R.B. — Anterograde Amnesia from Cardiac Arrest
History: R.B. was a 52-year-old man with a history of heart trouble who sustained a cardiac
arrest. Although his heart was successfully restarted, the period of anoxia (oxygen
deprivation) caused by the temporary halt in blood flow resulted in brain damage.
Deficit: The primary symptom of this brain damage was a PERMANENT ANTEROGRADE
AMNESIA, carefully documented by Zola-Morgan and his colleagues.
Death and Autopsy: Five years after the onset of amnesia, R.B. died of heart failure. His
family gave permission for histological examination of his brain.
Critical Finding: The investigators discovered that FIELD CA1 OF THE HIPPOCAMPAL
FORMATION was gone — its neurons had completely degenerated. Remarkably, this was
the ONLY significant damage found.
Conclusion: Loss of just the CA1 field of the hippocampus is sufficient to cause permanent
anterograde amnesia. Subsequent studies reported other patients with anterograde
amnesia caused by CA1 damage.

5.3 Why is Field CA1 So Vulnerable to Anoxia?

This is one of the most interesting questions in neuroscience. The answer has major implications:
7.
8.
9.
10.
11.

Important paradox: The very feature that makes CA1 so important for learning (its NMDA-receptor
richness and LTP capacity) also makes it the most VULNERABLE to metabolic disturbances.
6. ANTEROGRADE AMNESIA: FAILURE OF RELATIONAL
LEARNING

6.1 Clarifying the Nature of Declarative Memory

A common misconception is that declarative memories are simply 'verbal' memories — that
anterograde amnesia is merely an inability to learn new verbal information. This is INCORRECT.

Critical insight: Declarative memories are NOT necessarily verbal — they are RECOLLECTIONS
OF THINGS OR EVENTS. They represent a series of events, not a series of words.

Example: Memory of Morning Breakfast

Consider the author's memory of this morning's breakfast — putting on a robe, walking downstairs,
making coffee, making waffles. This memory:
• Contains MANY EVENTS, organized in TIME
• Is NOT verbal — it is perceptions of a series of EVENTS, not words
• Includes sensory information: sight of snow falling, smell of coffee, sounds of the grinder,
feel of warm slippers
• Is EPISODIC in nature — tied to a specific time and place

6.2 Episodic Memories: The Most Complex Declarative Memories

TERM DEFINITION
Episodic Memory Collections of perceptions of events organized in TIME, identified by
a PARTICULAR CONTEXT (time, place, circumstances). The most
complex form of declarative memory.
Semantic Memory General knowledge about the world — facts, concepts, meanings
— not tied to specific personal episodes.

• Episodic memories consist of COLLECTIONS OF PERCEPTIONS of events

• They are organized in TIME and identified by a particular CONTEXT
• They require the hippocampal formation to LINK TOGETHER the series of perceptions that
constitute a single episode
• People with anterograde amnesia CAN form perceptual memories of individual objects and
events...
• ...BUT their perceptual memories are ISOLATED — not tied together or to the context in
which they occurred
6.3 The Role of the Hippocampal Formation in Binding Memories
What does the hippocampal formation actually DO during an episode?

• It ties together a series of perceptions in such a way that their memories, too, are linked.
• It enables us to learn the RELATIONSHIP between stimuli that were present at the time —
the CONTEXT in which the episode occurred.
• Without it, seeing a particular person does not remind amnesic patients of other times they
have seen that person or of the things they have done together.

Conclusion: Anterograde amnesia appears to be a loss of the ability to learn ABOUT THE
RELATIONSHIPS AMONG STIMULI, including the TIME and PLACE in which they occurred and
the ORDER OF THEIR OCCURRENCE.

KEY CONCEPT: The Three Functions of the Hippocampus

• Consolidation: Converting short-term memories to long-term memories
• Context Binding: Linking stimuli to the context (time, place) in which they were
experienced
• Relational Learning: Encoding the relationships between stimuli present during an
episode
7. ROLE OF THE HIPPOCAMPAL FORMATION IN SPATIAL
MEMORY

Although spatial information need not be declared verbally (we can demonstrate our topographic
memories by successfully navigating), spatial memory is a prime example of relational learning — it
requires encoding the relationships between landmarks in the environment.

7.1 Human Evidence: H.M. and Spatial Memory

• Patient H.M. was never able to find his way around his present environment.
• After his parents moved following his surgery, he could never learn the layout of the new
home.
• People with anterograde amnesia are unable to consolidate information about the location of
rooms, corridors, buildings, roads, and other important items in their environment.

7.2 Clinical Case: Right Parahippocampal Cortex Damage (Luzzi et al.,

2001)

CASE STUDY: Patient with Right Parahippocampal Lesion

Lesion: A man with damage limited to the right parahippocampal cortex.
Deficit: He completely lost his ability to find his way around a NEW environment.
Compensatory strategies: The only way he could find his room was by COUNTING
DOORWAYS from the end of the hall or by seeing a RED NAPKIN that was located on top
of his bedside table.
Significance: Demonstrates that even unilateral (right-sided) damage to the hippocampal
formation or its associated cortices can produce significant spatial memory impairment.

7.3 Functional Imaging Studies: Navigation Activates the Hippocampus

London Taxi Drivers Study (Maguire, Frackowiak & Frith, 1997)
• London taxi drivers were asked to DESCRIBE THE ROUTES they would take when driving
from one location to another.
• A PET scan taken during their description of the route showed ACTIVATION OF THE
RIGHT HIPPOCAMPAL FORMATION.
• This is remarkable: simply thinking about navigation activates the spatial memory system.
Virtual Reality Navigation Study (Maguire et al., 1998)
• Subjects played a VIRTUAL REALITY COMPUTER GAME that permitted them to navigate
around a town.
• The subjects played long enough that the town's streets, buildings, open spaces, and other
features became familiar.
• Experimenters could close doors or put up barricades, requiring subjects to follow
ALTERNATIVE ROUTES.
• PET scanning while subjects navigated showed ACTIVATION OF THE RIGHT
HIPPOCAMPAL FORMATION.
• Critical finding: The AMOUNT OF ACTIVITY in this region was CORRELATED WITH THE
SUBJECTS' ACCURACY IN NAVIGATION.

Structural Changes in London Taxi Drivers (Maguire et al., 2000)

• London taxi drivers undergo EXTENSIVE TRAINING — approximately two years — to learn
to navigate efficiently in London.
• They receive their license only after passing a rigorous set of tests.
• Using MRI scans, Maguire et al. found that the volume of the POSTERIOR HIPPOCAMPUS
was LARGER in London taxi drivers than in control subjects.
• The volume of the ANTERIOR HIPPOCAMPUS was SMALLER in taxi drivers.
• Dose-response relationship: The LONGER an individual taxi driver had spent in this
occupation, the LARGER was the volume of the posterior hippocampus (and the smaller
was the anterior).
• Significance: This demonstrates that extensive spatial learning produces structural
changes in the human hippocampus — use-dependent neuroplasticity.

Place Cells: Neural Basis of Spatial Memory

Discovery: O'Keefe and Dostrovsky (1971) recorded the activity of individual pyramidal
cells in the hippocampus as animals moved around the environment.
Finding: Some neurons fired at a HIGH RATE only when the rat was in a PARTICULAR
LOCATION. Different neurons had different SPATIAL RECEPTIVE FIELDS.
Place Cell: A neuron that becomes active when the animal is in a particular location in the
environment; most typically found in the hippocampal formation. The region of the
environment into which entry produces increased firing is called the SPATIAL RECEPTIVE
FIELD.
Encoding: Information is represented by particular PATTERNS OF ACTIVITY in circuits of
neurons within the hippocampal formation, not by individual neurons encoding specific
locations.
In rodents: Most hippocampal place cells are found in the DORSAL hippocampus, which
corresponds to the POSTERIOR hippocampus in humans.
In monkeys: Most neurons encode information about what part of the environment the
animal is LOOKING AT rather than where it is located — called SPATIAL VIEW CELLS
(Rolls, 1996).
7.4 Place Cell Properties and Behavior
• A particular neuron might fire TWENTY TIMES PER SECOND when the animal is in a
particular location but only a FEW TIMES PER HOUR when elsewhere.
• Changes in environmental items AFFECT PLACE CELL FIRING and navigational ability.
• Maintaining positions: When experimenters MOVE THE STIMULI AS A GROUP, animals
simply REORIENT their responses.
• Disruption: When experimenters INTERCHANGE the stimuli so they are in a new
arrangement, the animals' performance AND place cell firing is DISRUPTED.

Clever Experiment: Skaggs and McNaughton (1998)

• Constructed an apparatus containing TWO NEARLY IDENTICAL CHAMBERS connected by
a corridor.
• Each day, rats were placed in the SAME CHAMBER; electrode clusters recorded
hippocampal place cell activity.
• Some place cells showed SIMILAR PATTERNS in each chamber, some showed
DIFFERENT patterns — suggesting the hippocampus 'recognized' both the similarities and
differences.
• On the last day, rats were placed in the OTHER CHAMBER. Initially, place cells responded
as if in the usual (north) chamber.
• Key observation: Once the rat saw the corridor and realized it was in the south chamber,
place cells IMMEDIATELY switched to the 'south' pattern — demonstrating rapid, context-
sensitive updating of spatial representations.
8. RELATIONAL LEARNING IN LABORATORY ANIMALS

The discovery that hippocampal lesions produce anterograde amnesia in humans stimulated
research into the exact role of the hippocampal formation in learning. Initially, animals with
hippocampal lesions appeared capable of learning most tasks — but these tasks primarily tested
stimulus-response (nondeclarative) learning. New paradigms requiring relational learning revealed
the same deficits as in humans.

8.1 The Morris Water Maze: A Classic Test of Relational Learning

Development and Design (Morris et al., 1982)
• Developed as a standard test of rodents' spatial abilities, now used widely in neuroscience
research.
• Consists of a CIRCULAR POOL, 1.3 meters in diameter, filled with water made opaque by
adding powdered milk.
• Water hides the location of a SMALL PLATFORM situated just beneath the surface of the
liquid.
• Rats are put into the water and swim until they encounter the hidden platform and climb onto
it.
• Released from a NEW POSITION on each trial.
• After a few trials, normal rats learned to swim DIRECTLY TO THE HIDDEN PLATFORM
from wherever they were released.

Why It Requires Relational Learning

• To navigate, animals must get their bearings from the RELATIVE LOCATIONS of stimuli
located OUTSIDE the maze (furniture, windows, doors).
• This requires encoding the SPATIAL RELATIONSHIPS between multiple environmental
cues — the hallmark of relational learning.
• Stimulus-response version: If animals are ALWAYS RELEASED FROM THE SAME
PLACE, they learn to head in a particular direction — this is non-relational and can be
solved without the hippocampus.

Results with Hippocampal Lesions

• Same starting position (non-relational): Rats with hippocampal lesions LEARN AS WELL
AS normal rats — confirming hippocampus is not needed for stimulus-response learning.
• Variable starting positions (relational): Hippocampally-lesioned rats SWIM IN AN
AIMLESS FASHION until they finally encounter the platform — demonstrating profound
impairment in relational spatial learning.
8.2 Hippocampal Lesions and Homing Pigeons (Gagliardo, Ioalè &
Bingman, 1999)
• Hippocampal lesions disrupted navigation in HOMING PIGEONS.
• Lesions did NOT disrupt the birds' ability to use the position of the SUN as a compass
pointing toward their home roost.
• What was disrupted: The ability to KEEP TRACK OF WHERE THEY WERE when they got
near the end of their flight — the time when birds normally use FAMILIAR LANDMARKS to
determine where they are.
• Conclusion: The hippocampus is critical for LANDMARK-BASED navigation (relational), but
not for sun-compass navigation (non-relational).

8.3 Hippocampal Volume and Spatial Ability: Cross-Species Evidence

A review by Sherry, Jacobs, and Gaulin (1992) found:
• The hippocampal formation of SPECIES OF BIRDS AND RODENTS that normally STORE
SEEDS in hidden caches and later RETRIEVE THEM is LARGER than that of animals
without this ability.
• This relationship between hippocampal size and spatial memory ability exists across many
species, providing strong evolutionary evidence for the hippocampus's role in spatial and
relational memory.

8.4 Role of Long-Term Potentiation (LTP) in Relational Learning

In the first part of this chapter, we saw how synaptic connections could be quickly modified in the
hippocampal formation, leading to long-term potentiation (LTP) or long-term depression (LTD). Are
these changes related to the role the hippocampus plays in learning?

Knockout Mouse Studies: McHugh et al. (1996) and Tsien, Huerta & Tonegawa
(1996)
• Researchers developed TARGETED MUTATIONS of the gene responsible for production of
NMDA receptors.
• Mutation affected ONLY THE CA1 PYRAMIDAL CELLS — in all other parts of the brain,
receptors were normal.
• Result 1: Lack of NMDA receptors PREVENTED establishment of long-term potentiation in
field CA1.
• Result 2: Although CA1 cells still showed spatial receptive fields, these fields were
LARGER and LESS FOCUSED than in normal animals.
• Result 3: Knockout mice learned a MORRIS WATER MAZE MUCH MORE SLOWLY than
mice with normal CA1 NMDA receptors.
'Smart Mouse' Study: Tang et al. (1999)
• Performed a genetic manipulation that caused INCREASED PRODUCTION of the NMDA-
R2B subunit in the forebrain.
• NMDA receptors containing the R2B subunit produce a SLIGHTLY LARGER EXCITATORY
POSTSYNAPTIC POTENTIAL.
• As a consequence, LONG-TERM POTENTIATION WAS ENHANCED in hippocampal slices
from the genetically modified mice.
• These animals LEARNED TO FIND THE PLATFORM in a Morris water maze FASTER than
normal animals.
• Conclusion: Hippocampal LTP PLAYS AN IMPORTANT ROLE IN RELATIONAL
LEARNING. Enhancing LTP improves spatial learning; disrupting it impairs spatial learning.

CASE STUDY: The 'Smart Mouse' — Important Caveats

The Tang et al. (1999) study suggested a genetic manipulation might enable us to produce
smarter animals or even smarter people. However, some important caveats apply:
Natural selection argument: If having more NMDA-R2B subunits were entirely beneficial,
natural selection would probably have bequeathed them to us already.
CA1 vulnerability: The large concentration of NMDA receptors in CA1 makes these
neurons PARTICULARLY SUSCEPTIBLE TO DAMAGE by metabolic disturbances. The
very feature that makes them good learners makes them fragile.
Trade-off: The flexibility that CA1's NMDA-rich neurons provide for rapid learning
undoubtedly contributes to our learning ability — but also renders these neurons particularly
vulnerable to seizures, anoxia, or hypoglycemia.
9. COMPREHENSIVE SUMMARY AND EXAM GUIDE

9.1 Master Concept Map: Key Relationships

Brain Structure Memory Function Effect of Damage Key Evidence
Hippocampus (CA1) Consolidation of STM Anterograde amnesia H.M., Patient R.B.
to LTM; relational
learning
Entorhinal Cortex Main input/output Anterograde amnesia Connectivity studies
gateway to
hippocampus
Perirhinal Cortex Relay between Memory impairment Lesion studies
entorhinal cortex and
association cortex
Parahippocampal Relay and spatial Spatial memory Luzzi et al. 2001
Cortex processing deficits
Right Hippocampus Spatial/navigational Navigation deficits Maguire et al., taxi
memory drivers
Amygdala Emotional influence Conditioned Conditioning studies
on memory reinforcement
disrupted
Nucleus Accumbens Reward/reinforcement Disrupted Dopamine infusion
(via dopamine) instrumental studies
conditioning

9.2 Common Exam Questions and How to Answer Them

Q1: What is anterograde amnesia and how does it differ from retrograde
amnesia?
Answer framework: Define both terms, highlight the time-frame difference (after vs. before brain
damage), then discuss the case of H.M. as the prime example. Note that pure anterograde amnesia
is rare — usually some retrograde amnesia is also present.

Q2: Describe the spared learning abilities in anterograde amnesia with

examples.
Answer framework: State that perceptual, stimulus-response, and motor learning are spared.
Provide one specific experiment for each type: broken drawings (perceptual), eyeblink conditioning
(S-R), button-press sequence (motor). Emphasize that in all cases, patients learned normally but
had NO DECLARATIVE MEMORY of having learned.

Q3: Distinguish between declarative and nondeclarative memory with

examples.
Answer framework: Define both types. Use the bicycle example. Discuss the Graf, Squire &
Mandler (1984) experiment showing explicit recall was impaired but implicit word completion was
intact. Include the table of memory tasks.

Q4: What is the role of the hippocampal formation in spatial memory?

Answer framework: Discuss Morris water maze, London taxi driver studies (both MRI volume and
PET activation), place cells (O'Keefe and Dostrovsky), and the Skaggs & McNaughton experiment.
Connect to the broader theme of relational learning.

Q5: Why is field CA1 particularly vulnerable to anoxia?

Answer framework: Explain the NMDA receptor richness of CA1, the mechanism (excessive
glutamate → NMDA activation → calcium entry → neuronal death), Patient R.B. as clinical
evidence, and the important paradox that this same feature makes CA1 an excellent site for LTP
and learning.

Q6: What is the role of LTP in relational learning?

Answer framework: Define LTP, explain its dependence on NMDA receptors in CA1. Discuss the
knockout mouse study (no LTP = slower maze learning, less focused place fields), and the Tang et
al. smart mouse study (enhanced LTP = faster maze learning). Conclude that hippocampal LTP is
integral to relational learning.

9.3 Critical Terms: Complete Glossary

TERM DEFINITION
Anterograde amnesia Difficulty in learning new information following brain damage;
impaired consolidation of new declarative memories.
Retrograde amnesia Inability to remember events that preceded brain damage.
Consolidation The process by which short-term memories are converted into long-
term memories.
Short-term memory Immediate memory for events, maintained by rehearsal; limited
capacity; H.M. had intact STM.
Long-term memory Relatively stable memory of events that occurred in the more
distant past; requires consolidation.
Declarative memory Memory that can be verbally expressed; memory for events/facts in
a person's past; explicitly available to conscious recollection.
Nondeclarative memory Memory whose formation does not depend on the hippocampal
formation; includes perceptual, stimulus-response, and motor
memory.
Episodic memory Memory for collections of perceptions of events organized in time,
identified by a particular context.
Relational learning Learning the relationships among multiple stimuli, including time
and place of occurrence.
Korsakoff's syndrome Permanent anterograde amnesia caused by brain damage resulting
from chronic alcoholism or malnutrition.
Hippocampal formation A brain structure critical for consolidation of declarative memories
and relational learning.
Perirhinal cortex Region of limbic cortex adjacent to hippocampal formation; relays
information between entorhinal cortex and other brain regions.
Parahippocampal cortex Region of limbic cortex adjacent to hippocampal formation; relays
information between entorhinal cortex and other brain regions.
Place cell A neuron in the hippocampal formation that becomes active when
the animal is in a particular location in the environment.
Spatial receptive field The region of the environment into which the entry of an animal will
produce an increase in the firing rate of a place cell.
Long-term potentiation A long-lasting increase in synaptic strength following high-frequency
(LTP) stimulation; believed to be the cellular mechanism of learning.
Field CA1 A sub-region of the hippocampus, particularly rich in NMDA
receptors, critical for LTP and anterograde memory formation;
highly vulnerable to anoxia.
NMDA receptor A type of glutamate receptor that requires both ligand binding AND
membrane depolarization to open; key to LTP induction.
Morris water maze A behavioral task requiring rats to find a hidden platform in a pool
using distal visual cues; requires relational spatial learning.
Bilateral medial temporal Surgical removal of both medial temporal lobes; the procedure that
lobectomy caused H.M.'s anterograde amnesia.
10. ADDITIONAL CASE STUDIES FOR EXAM PREPARATION

CASE STUDY: Patient H.M. — Complete Summary

Full name: Henry Molaison (1926-2008). Referred to as H.M. to protect his privacy during
his lifetime.
Pre-surgical condition: Extremely severe epilepsy that could not be controlled even by
high doses of anticonvulsant medication. Epilepsy caused by a head injury received when
struck by a bicycle at age nine (Corkin et al., 1997).
Surgery: Bilateral removal of the medial temporal lobe in 1953 by surgeon William Beecher
Scoville. Thirty operations had been performed on psychotic patients before; it was not until
H.M.'s operation that anterograde amnesia was discovered.
Why H.M. was different: H.M. was reasonably intelligent and not psychotic, so his
postoperative deficit was discovered immediately.
Memory profile: Could remember events from early in his life; intact STM; PROFOUND
anterograde amnesia (could not form any new declarative memories after 1953).
Daily life: Could not remember his doctors, the hospital, what he had eaten for breakfast, or
even recognize people he had met many times since his surgery.
Impact on neuroscience: Led to the discovery that bilateral hippocampal damage causes
anterograde amnesia, changed neurosurgical practice, and established the conceptual
distinction between declarative and nondeclarative memory.

CASE STUDY: Patient R.B. — Evidence for CA1's Critical Role

Background: 52-year-old man with heart disease.
Event: Cardiac arrest → temporary anoxia → brain damage → permanent anterograde
amnesia.
Post-mortem finding: Complete degeneration of field CA1 of the hippocampal formation
with minimal damage elsewhere.
Significance: Provides the clearest evidence that damage to the hippocampal formation
produces anterograde amnesia, and that field CA1 is the critical structure. Subsequent
studies confirmed CA1 damage in other amnesic patients with similar histories.

CASE STUDY: London Taxi Drivers — Use-Dependent Hippocampal Plasticity

Background: London taxi drivers must learn the layout of an enormously complex city.
Training takes ~2 years; they receive their license only after rigorous testing.
MRI study (Maguire et al., 2000): Volume of POSTERIOR hippocampus was LARGER in
taxi drivers than controls. Volume of ANTERIOR hippocampus was SMALLER.
Dose-response: The longer an individual had been a taxi driver, the larger was the
 posterior hippocampus (and the smaller the anterior hippocampus).
Navigation PET study (Maguire et al., 1997): When taxi drivers described complex routes,
the right hippocampal formation became activated.
Theoretical significance: Demonstrates that extensive spatial learning produces
STRUCTURAL CHANGES in the human hippocampus — providing in vivo evidence for
experience-dependent neuroplasticity in a region critical for relational memory.

CASE STUDY: The 'Smart Mouse' (Tang et al., 1999) — LTP and Relational
Learning
Genetic manipulation: Increased production of NMDA-R2B subunit in the forebrain of
mice.
Cellular effect: NMDA receptors with R2B subunit produce slightly larger EPSPs → LTP
was enhanced in hippocampal slices.
Behavioral effect: Genetically modified mice learned the Morris water maze platform
location FASTER than normal mice.
Implication: Strongly suggests that hippocampal LTP plays an important role in relational
learning — the faster and stronger the LTP, the faster the spatial learning.
Paradox: More NMDA receptors = better learner BUT also more vulnerable to excitotoxic
damage. Natural selection has not maximized NMDA receptors, suggesting there is an
optimal level that balances learning ability against vulnerability.

11. EXAM WRITING STRATEGY

Tips for High-Score Exam Answers on This Unit

12.
13.
14.
15.
16.
17.
18.
Model Answer Structure for 15-20 Mark Questions
Section Content Marks (Approximate)
Introduction Define the key terms in the 2-3 marks
question; state the major brain
structures involved
Main Body Part 1 First major concept with 4-5 marks
definition, mechanism,
experimental evidence
Main Body Part 2 Second major concept or 4-5 marks
contrasting idea with evidence
Case Studies Named patient (H.M., R.B.) or 3-4 marks
animal study with specific
findings
Synthesis/Conclusion Link the concepts back to the 2-3 marks
broader principle of relational
learning

— END OF UNIT 3 NOTES —

Source: Foundations of Physiological Psychology, 6th Edition, Chapter 12