C ha p t er

1
Introduction
The Basics of Psychological Learning
and Memory Theory

I. Introduction
A. Categories of Learning and Memory
B. Memory Exhibits Long-Term and Short-Term Forms
II. Short-Term Memory
A. Sensory Memory and Short-Term Storage
B. Working Memory
C. The Prefrontal Cortex and Working Memory
D. Reverberating Circuit Mechanisms Contrast with Molecular Storage Mechanisms for Long-Term Memory
III. Unconscious Learning
A. Simple Forms of Learning
B. Unconscious Learning and Unconscious Recall
C. Unconscious Learning and Subject to Conscious Recall
D. Operant Conditioning
E. Currently Popular Associative Learning Paradigms
IV. Conscious LearningSubject to Conscious and Unconscious Recall
A. Declarative Learning
B. Spatial Learning
V. Summary
Further Readings
Journal Club Articles
References

I. Introduction An understanding of the cellular and molecular

basis of learning and memory of course requires a firm
Knowledge is power and learning is the tool we foundation in understanding the behavioral processes
use to get it. For that reason humans have evolved these mechanisms subserve. This first chapter serves
extremely sophisticated mechanisms for learning as an introduction to the basics of learning and mem-
new information and storing it for subsequent recall. ory, its theory and terminology. This will provide you
This book will be a description of recent laboratory with the fundamental terms most psychologists use to
discoveries that have begun to scratch the surface of describe the types and forms of learning and memory
the amazingly complex phenomenon of learning and that we will be discussing throughout the book.
memory, focusing on their cellular and molecular What is learning? Before we can begin to effectively
bases. discuss categorizing types of learning and memory, it


 1. Introduction

is useful to define both of the terms we will be using output on the part of the person involved. For exam-
extensively throughout this book: learning and ple, an experimenter would have to have them
memory. Both of these terms are so widely used respond with David instead of I dont know
and implicitly understood that there is a great temp- when they showed them my picture. Nevertheless, it
tation to say learning is when you learn something is important to remember that this definition is based
and memory is when you remember it. This type of in experimentation, not theory.
definition obviously is not going to take us very far. At the other end of the spectrum is the criticism
Upon serious reflection it becomes clear that nei- that the definition is too broad. It certainly covers
ther learning nor memory is easy to define, and many types of alterations in behavior, such as sim-
indeed learning and memory psychologists con- ple sensitization and habituation, which most people
tinue to debate these definitions to this day. In this would not consider as real learning (this is illus-
book we will define learning as: the acquisition of an trated in Box 1, for example). Nevertheless, a consid-
altered behavioral response due to an environmental erable body of literature is available indicating that
stimulus. In other words, learning is when an animal many simple forms of behavioral modification qualify
changes its behavior pattern in response to an experi- as learned responses, and most researchers in the field
ence. Note that what is defined is a change in a behav- agree with this. These forms of simple, non-associative
ior from a pre-existing baseline. Dont get confused: learning are described in Section III of this chapter,
learning is not a response to an environmental stimu- and in more detail in Chapter 3 of this book.
lus, but rather is an alteration in that response due to
an environmental stimulus. An animal has a baseline
response, experiences an environmental signal, and
A. Categories of Learning and Memory
then has an altered response different from its previ-
ous response. This is learning (see Figure 1). This broad, umbrella-like definition of learning
Memory is defined as the storage of the learned covers so many different types of behavioral modi-
item, which of course must be subject to recall by fications that some sort of organizing principle and
some mechanism. attendant nomenclature are called for. We will use
These definitions are functional definitions that an organizational framework developed and pro-
lend themselves to experimental application. An mulgated by Larry Squire and Eric Kandel (13). As
experimentalist has to be able to observe something a starting point we will use their system, and I would
(and ideally measure it) in order to be able to test a be remiss if I did not credit their many significant and
hypothesis. The definitions of learning and memory influential contributions in this area.
that are used in this book derive directly from the In this scheme human memory is typically divided
experimentalist mindset. This practical orientation is into declarative and non-declarative types, also
both a strength and a weakness for the definitions known respectively as explicit and implicit memory
their ready application in practice leads to limitations (see Figure 2). This type of system, subdividing mem-
for their use in theory. ory into several separately identified components, dis-
For example, one criticism of this definition of tills the modern concept of multiple memory systems.
learning is that it is too narrow. If someone learns my It is now clear that different anatomical structures in
name and stores it as a perfectly legitimate memory, the brain are involved in different types of memory
that learned item may never be manifest as an altered formation. Moreover, the different systems can oper-
behavioral output on their part. This is a completely ate as parallel processors, operating independently.
valid theoretical criticism and a limitation to the defi- This allows multi-tasking, with conscious and uncon-
nition. The rebuttal to this argument is that in order scious memory systems operating simultaneously and
for one to ever prove that such a memory exists, one increasing the overall memory throughput of the
would have to demonstrate an altered behavioral CNS. Figure 2 briefly summarizes the major subdivi-
sions of human memory, along with the associated
known areas of the CNS that are involved in those
Learning: The acquisition of an altered behavioral specific types of memory. We will discuss most of the
response due to an environmental stimulus. major subdivisions listed in Figure 2 in greater detail
later in this chapter, and in Chapter 3 of this book.
Memory: The processes through which learned information is
stored.
The multiple memory systems concept is impor-
tant and soundly based on functional neuroanatomy.
Recall: The conscious or unconscious retrieval process However, a different, cognitively based framework is
through which this altered behavior is manifest.
also useful to consider. This additional system is based
Figure 1 Definitions of learning, memory, and recall. on whether different types of learning and memory are
 I. Introduction 

Box 1

Learning in a plant? sensitization

in the venus flytrap

Our functional definition of learning is: a change in has evolved a mechanism whereby stimulation of a
an animals behavioral responses as a result of a unique single trigger hair is insufficient to cause closure of the
environmental stimulus. This broad definition is useful trap. Two hairs must be stimulated in succession (or
in that it encompasses various non-associative forms simultaneously) to trigger a trapping response. Thus,
of learning such as sensitization and habituation, but in one circumstance stimulating a particular trigger hair
the breadth of the definition can be criticized. This can will give no response, whereas depending on recent his-
be illustrated by consideration of sensitization in the tory stimulating the same trigger hair will in another
Venus flytrap plant. instance give trap closure. This is clearly an example
Although plants are not thought of expressing of an altered response that depends on a prior environ-
behavior in the same sense as animals, plants can and mental stimulus. In a sense, the mechanical stimulation
do respond to environmental stimuli. We are all famil- of the first trigger hair could be viewed as analogous to
iar with the phototactic responses of plants as they turn sensitizing the plant, in order that it respond to the
to follow the sun, foliage changes in response to cool- mechanical stimulation of the second hair. Venus flytrap
ing weather, and the nocturnal closing of certain flow- photograph by Muriel Weinerman.
ers, just to name a few simple examples. However, these
types of responses are really more akin to reflexive, non-
learned behaviors in animals.
One intriguingly complex, multi-component response
of a plant to an environmental stimulus is exhibited by
Dionaea muscipula, commonly known as the Venus fly-
trap. This carnivorous plant, indigenous to the peat
bogs of the Carolinas in the southeastern United States,
supplements its nutrition by capturing and digesting
insects. Insects are trapped by Dionaea when they land
in one of the plants V-shaped leaves, which closes on
the hapless victim like a miniature steel bear trap.
It is the triggering mechanism for closure of the trap
that warrants our attention. Each half of the V-shaped
trap has on its inward facing surface three trigger hairs.
Mechanical stimulation of these hairs is what elicits
closure of the trap. To eliminate false alarms, Dionaea

consciously or unconsciously processed. Thus, using illustrative purposes, and for the rest of this chapter
this system one can divide learning into two broad and in Chapters 2 through 6 we will cover many spe-
classesunconscious learning and conscious learning. cific examples in each category.
For the purposes of this framework we also introduce The nomenclature summarized in Figure 3 empha-
a recall term (see Figure 3), and apply conscious and sizes that any given memory event is comprised of
unconscious to it as well. Thus, any type of memory three components: learning; storage; and recall. An
(with one exception, see below) falls into one of four item or event is learned, stored for some period of time,
categories: unconscious learning with unconscious and recalled. Highlighting these three components is
recall; unconscious learning subject to conscious recall; necessary, because each corresponds to a distinct mole-
conscious learning subject to unconscious recall; and cular and cellular set of events.
conscious learning subject to conscious recall. Specific It is also important to note that the category for
examples of each category are listed in Figure 3 for the learning, memory, and recall of a specific bit of
 1. Introduction

HUMAN MEMORY

DECLARATIVE NON-DECLARATIVE
(EXPLICIT) (IMPLICIT)

FACTS EVENTS

PROCEDURAL PRIMING SIMPLE NON-ASSOCIATIVE

(SKILLS AND CLASSICAL LEARNING
HABITS) CONDITIONING

EMOTIONAL SKELETAL
RESPONSES MUSCULATURE

MEDIAL STRIATUM NEOCORTEX AMYGDALA CEREBELLUM

TEMPORAL REFLEX
LOBE PATHWAYS

Figure 2 Subdivisions of human memory and associated brain regions. Human memory is typically divided into declarative and non-
declarative types, also known as explicit and implicit memory, respectively. In addition to various types of memory described in the text,
priming is also listed. Priming is unconscious memory formation. An example of priming is if one hears or reads a word, for a period of time
afterward one is more likely to use that word in conversation or in a word completion task. This occurs even if no conscious memory for
having heard the word is formed. Chart adapted from Milner, Squire, and Kandel (13).

Hierarchical Organization of Memory

Unconscious learning Conscious learning Working memory

Storage Storage
(unconscious) (unconscious)
Conscious
storage and
conscious recall
Subject to conscious
Subject to conscious Unconscious recall
recall recall

Declarative learning
Spatial learning
Trace conditioning Conscious associative
Operant conditioning conditioning
Hippocampus-
dependent
contextual fear
conditioning
Taste learning
Conditioned taste Non-associative learning Associative learning Motor learning
aversion Habituation Pavlovian
Sensitization conditioning
Dishabituation Delay Eye-blink
conditioning
Cued Fear
conditioning

Figure 3 Hierarchical organization of memory. Short-term and long-term memory is subject to being learned by either conscious or
unconscious processes. Similarly, memory can be recalled either consciously or unconsciously. Many forms of simple learning such as motor
learning, simple associative conditioning, and non-associative learning can be learned and recalled unconsciously. More complex forms of
learning typically involve conscious processes. Short-term working memory is listed as a separate category because it is essentially entirely
conscious and not stored for more than a few seconds.
 I. Introduction 

information is not static over time, but subject to either short-lasting or long-lasting. With only a few
change. This can be illustrated by considering the exceptions (see Box 2), the duration of the memory for
learning and recollection of a phone number that a learned event depends on the number of times an
becomes familiar with repetition. One first looks animal experiences a behavior-modifying stimulus.
up the number and consciously stores and recalls For example, a single repetition (or training trial)
the number. Over time one repetitively punches in may elicit a memory that lasts only a few minutes,
the number and it is subject to being learned uncon- whereas repeated stimulations will likely result in
sciously as a motor pattern, and recalled uncon- memory lasting hours to days. Repeated presentations
sciously in the same way. This is one example of how of multiple training trials can elicit memory lasting for
the same bit of information, over time, can be subject even more prolonged periods, up to the lifetime of the
to conscious learning, unconscious learning, conscious animal. Thus, the acquisition of memory is a graded
recall, and unconscious recall. phenomenon (see Figure 4).
Finally, note that storage is unconscious in this One exciting area of contemporary learning research
model. This emphasizes the underlying nature of is to try to understand the basis for this attribute. It is
the storage mechanismsthey do not require ongo- intriguing to wonder how repeated presentations of the
ing conscious rehearsal. This has critically important identical environmental stimulus can uniquely elicit a
implications concerning the cellular and molecular long-lasting behavioral alteration, especially when one
processes that underlie memory storage. They must be considers that the behavioral output (e.g., enhanced
stable and capable of self-perpetuation in the absence responsiveness) is identical in the short- and long-lasting
of ongoing conscious input. forms. This is still fairly mysterious at present for the
This is not to say that all forms of memory are various mammalian systems that we will be discussing;
stored unconsciouslyclearly several forms of short- however, significant progress has been made address-
term working memory are conscious. A good exam- ing this issue in the Aplysia invertebrate model system
ple of this is short-term storage of a phone number, that will be discussed in Chapter 3.
where one can store information over time essen- Long-term memory also has the general attribute
tially by conscious repetition over a given time span. that it undergoes a period of consolidation. Decades
However, this form of memory is in a separate cate- ago it was discovered that, for a period of time after
gory from longer-term forms of memory from a cellu- the training period, generally on the order of hours,
lar and molecular perspective (see Figure 3). Working memories that were normally destined to become
memory can be stored as a short-term change in fir- long-term memories were susceptible to disruption.
ing pattern in cortical neurons, for example in a Disruption of nascent long-term memories can be
reverberating circuit. As such, it does not require any brought about by trauma, for example, or in a more
persisting biochemical modification for its mainte- refined manipulation application of inhibitors of
nance. Indeed, at the molecular level this seems likely protein synthesis can block memory consolidation
to be the distinguishing characteristic of working (Figure 5). Thus it is clear that some set of molecular
memory. It is memory that cannot sustain itself in the processes is occurring for some period of time after
absence of continuing neuronal firing. the training trial, which are necessary for memory to
These categories of learning and memory roughly be established as truly long-lasting. Once the critical
correspond to the typically used non-declarative time window has passed, the same disruptive manip-
memory and declarative memory nomencla- ulations have no effect on memory storage. Studies
ture popularized by Squire and Kandel (Figure 2), of the cellular and molecular mechanisms contribut-
and widely accepted and utilized. I also emphasize ing to the consolidation of long-term memory will be
the conscious/unconscious terminology because it an area of emphasis in Chapters 2, 3, 6, and 10 of this
highlights the cognitive differences between the two book.
forms. Most importantly, this terminology semanti- There has been a resurgence of interest in the
cally separates the learning from the memory storage consolidation phenomenon lately because several
from the recallan important mindset to adapt as we groups have reported that previously stored memo-
seek to understand learning and memory events in ries are subject to disruption in certain circumstances.
molecular terms. Specifically, for some types of memory an event
already learned and stored in long-term memory is
B. Memory Exhibits Long-Term and selectively subject to disruption when it is recalled.
The basic experimental observation is that while pro-
Short-Term Forms
tein synthesis inhibitors do not wipe out stored mem-
Emphasized by Eric Kandel, Jim McGaugh, and ory, the same protein synthesis inhibitor treatment
many others (4), almost all forms of memory can be will disrupt memory if the subject is simultaneously
 1. Introduction

Box 2

Non-graded acquisition of memoryfood aversion

and imprinting

Generally, if an animal consumes a novel foodstuff that

subsequently causes sickness, even after a single such
experience the animal will exhibit a life-long aversion
to that particular food. While for animals in the wild
the survival value of this type of learning is obvious,
the phenomenon can have unintended consequences.
For example, I once got food poisoning after eating a
bowl of New England clam chowder; to this day even
the sight of a can of New England clam chowder on
the grocery store shelf is enough to send me scurrying
to the next aisle. This is a textbook case of conditioned
food avoidancebeing from Alabama, I had never had
clam anything until that day. I certainly will fastidiously
avoid future clam encounters of any kind.
While I have not personally experienced it, hatchling
chicks exhibit a robust form of learning termed imprint-
ing. A newborn bird will develop a strong, long-lasting
affinity for whatever it sees in the first hour after hatch-
ing. In one famous example, a group of young geese
imprinted on the experimental ethologist Konrad Lorenz.
In experimental situations chicks will even imprint on
inanimate objects, such as red boxes or dolls. Of course,
While most forms of long-term memory exhibit in the wild this type of learning serves a useful purpose,
graded acquisition, some types of learning are so criti- as hatchlings will almost always first see their mother
cal to an animals survival that extremely robust learn- and imprint upon her. The chicks will then stick close by
ing mechanisms have evolved to subserve them. One the mother as she guides and protects them through the
striking example of this is conditioned food avoidance. perilous fledgling period.

required to recall the information (56). Thus, pair- (the engram), changing them at least transiently and
ing protein synthesis inhibitors with a behavioral task triggering a new process of memory reconsolidation.
requiring information recall can lead to a selective Finally, to round out our terminology we need to
loss of a previously stored memory from long-term introduce three terms related to the loss or suppres-
stores. This intriguing process is referred to as recon- sion of memories: extinction; forgetting; and latent
solidation of memory. The necessity for a process of inhibition. Forgetting is woefully familiar to most of
memory reconsolidation highlights the fact that previ- us, and its basis is essentially unexplored. At a min-
ously formed, apparently stable, memories are labile imum it can be defined as a failure over time of the
after recall, and must be restabilized for continued storage or recall processes.
storage. Extinction is the specific erasure of a previous
The process of memory reconsolidation also illus- memory in response to a new environmental stimu-
trates that recall is its own unique process; recall is lus. Extinction has largely been studied in the context
not simply a passive process that does not impact the of reversal of learning. For example, if your cafeteria
underlying memory storage mechanism. Rather, recall, serves hamburgers every Monday you will learn over
in at least some instances, directly impacts the molecu- time that the cafeteria always serves hamburgers
lar and cellular processes underlying memory storage on Monday. If at some later point they stop serving
 I. Introduction 

1000 Training Testing at 24 Hrs.

4 trains / day 100
for 4 days

latency to step-down (s)

80
Duration of withdrawal (% of control)

60

40

500 20
*
0
4 single shocks
Control Inhibitor Control Inhibitor

Single Figure 5 Protein synthesis inhibitors block consolidation of

tail shock long-term memory. Inhibitors of protein synthesis typically block
the ability of learned information to be consolidated into a long-
100 lasting form. In this experiment rats were trained in a step-down
Control avoidance paradigm (see Chapter 4). Animals are placed on an
0 elevated platform in the middle of an electric grid and receive a
0 1 4 7 mild foot shock when they step down from the platform. On the
Days after training training day animals that received a saline infusion (CONTROL)
or the protein synthesis inhibitor anisomycin (INHIBITOR) both
Figure 4 Graded acquisition of memory. Multiple training trials quickly step down from the platform (latency to step-down, Y-axis).
typically result in more robust and long-lasting memory formation.
Twenty-four hours later the control animals exhibit a much longer
In this case sensitization of the gill-withdrawal reflex in Aplysia califor-
latency to step down, indicating that they have learned to avoid
nica was measured by quantitating the duration of gill withdrawal in
the electrified floor. Animals treated with protein synthesis inhibi-
response to a slight touch (duration of withdrawal, Y-axis). Delivery
tor have not consolidated their memory for the step-down training,
of a tail shock to the animal elicits sensitization, and an increase in
and exhibit a short latency to step down just as they did on the first
the magnitude of the protective gill withdrawal reflex (see Box 3 and
day. Additional experiments (not shown) have demonstrated that
text). Increasing numbers of training trials (tail shocks) increases both
anisomycin treatment immediately after training is also effective at
the duration of the memory (days of duration) and the magnitude of
blocking memory consolidation, indicating that consolidation is a
the learned response. Adapted from Kandel (14).
post-training phenomenon (5).
A few words about the particulars of the Aplysia model system
are appropriate at this point, although we will discuss this system
in much greater detail in later parts of the book. Much (but by no
means all) of the work in Aplysia has been geared toward under-
standing the basis of sensitization in this animal. Aplysia has on is an example of extinguishing a previously learned
its dorsum a respiratory gill-and-siphon complex, which is nor- response. This is an extinction of a memory that
mally extended when the animal is in the resting state. If the gill Monday means hamburgers. Similar to forgetting, the
or siphon is lightly touched (or experimentally, squirted with a
unique mechanisms underlying extinction have not
Water-Pic), this elicits a defensive withdrawal reflex in order to
protect the gill from potential damage. This defensive withdrawal been extensively studied. One intriguing speculation is
reflex can undergo both habituation (by repeated light stimuli) that the reconsolidation mechanism may be involved
and sensitization. Sensitization occurs when the animal receives in some cases, the thinking being that perhaps recon-
an aversive stimulus, for example a modest tail-shock experimen- solidation is the process that has evolved to allow spe-
tally or a predatory nip in the wild (see Box 3). After sensitizing
cific erasure of previously learned material, by opening
stimulation, the animal exhibits a more robust, longer-lasting gill-
withdrawal in response to the identical light touch or water squirt. up a period of susceptibility on recall (56).
Acquisition of this sensitization response is graded; repetitive Latent inhibition is the mirror image of extinction.
sensitizing stimuli can give sensitization lasting minutes to hours Latent inhibition refers to the capacity of prior experi-
(one to a few shocks), or weeks (repeated training trials over a few ence to suppress (inhibit) new learning. The latent
days). We will return to the Aplysia system in later chapters of the
in latent inhibition refers to the attribute that the pro-
book, where we will discuss several of the biochemical mechanisms
underlying the short- and long-term modification of this behavioral cess is passive and not generally recognizable until
response. one observes a failure of learning. Latent inhibition can
be illustrated by the following example. Over a life-
time of food consumption you passively and uncon-
hamburgers on Monday it will take a while to relearn sciously learn a wide variety of tastes. Familiar tastes
that contingency. Over time, you will no longer from foods that you have repeatedly consumed are
assume that if its Monday that means hamburgers, not subject to conditioned food avoidance (see Box 2)
and similarly will no longer infer that if they are serv- if they are paired with a nausea-inducing agent. The
ing hamburgers then it is Monday. This disassociation prior experience with the familiar taste leads to latent
10 1. Introduction

inhibition of subsequent aversive conditioning; having Short-term storage refers to retention of information
a latent memory that the taste has not been previously in the short-term system after the information has been
associated with malaise leads to an inhibition of the processed and has reached consciousness. The process-
formation of a new, different association. ing may have been either the processing of new sen-
sory input, or processing in the sense of recalling a
previously stored memory, hence short-term storage
operates on both new and old information (Figure 6A).
II. Short-term memory In the case of handling new information, short-term
storage may operate as a step in the sequence of events
The quickest, earliest stages of memory of necessity leading to long-term storage of that information.
deal with processing transient sensory and perceptual If a person is distracted, information is rapidly lost
stimuli. The buffers for holding onto sensory informa- from short-term storage. One commonly-used tech-
tion for seconds or a few minutes after their termina- nique to counteract this fact (in humans at least) is
tion in the environment are referred to as short-term ongoing repetition or rehearsal of the information held
memory. The short-term memory system is divided into in short-term storage (Figure 6A). As a first approxi-
three basic components: sensory memory; short-term mation, the information in your short-term stor-
storage; and working memory, each with different age is the information of which you are consciously
functions. aware.
It is important to realize that short-term memory is
bidirectional. It is clear that short-term memory deals
with sensory perceptions as already mentioned, but B. Working Memory
short-term memory also handles information that is It is possible to hold a fact in short-term storage
recently recalled from long-term stores. Thus, short- without doing anything with it. However, if the infor-
term memory is both an input device and an output mation is manipulated and further processed in any
device. It not only handles new information freshly way, it is referred to as being held in working memory.
perceived, it also handles old information freshly Thus, the term working memory refers specifically to
recalled. Old information must be brought forward the type of memory system used to hold information
into a short-term memory store for utilization, and for short periods of time while it is being utilized. A
this is also a component of short-term memory. simple example is doing arithmetic calculations using
remembered numbers (what is 456?). Mentally
multiplying 456 is clearly a different memory task
A. Sensory Memory and Short-Term Storage
than simply remembering the number 224 for a few
The first component of the short-term memory sys- seconds.
tem deals exclusively with freshly perceived informa- Alan Baddeley has presented a refined model of the
tion, for example the face of someone you have just working memory component of short-term memory
met. The sensory input (visual in this example) begins that is a significant addition to the simpler multi-store
its journey into memory by passing into the first stage model presented in the previous section and in Figure
of the short-term memory system. This initial, tran- 6A. In the Baddeley model, the passive sensory reg-
sient stage of sensory information storage is referred isters and short-term stores (Figure 6A) are also aug-
to as sensory memory, or the sensory register (Figure mented by a working memory module (Figure 6B).
6A). While it is difficult to define exactly when per- In the Baddeley module, three different storage sys-
ception ceases and short-term memory takes over, it tems contribute to the working memory component of
is clear that sensory input, be it touch, taste, smell, short-term memory. The phonological loop is respon-
sight or sound, must pass into a short-term store sible for short-term storage of auditory and spoken
in order to be further processed as part of a lasting language information. Limitations to the capacity of
memory. the phonological loop are responsible for the familiar
The sensory register is the first stage of processing limits on digit-span memory capacity, for example.
new information into a memory. Presumably, each The visuospatial sketchpad is conceptually similar to
different sensory system has dedicated components the phonological loop, except that it deals with visual
of the sensory register that contribute to passing its and spatial information. The episodic buffer is the
unique information along to memory. However, two component that deals with holding and manipulating
sensory registers have been widely studied, and have information recently recalled from long-term storage.
been poetically named. Echoic memory refers to the These three components are regulated by a central
auditory sensory store, while iconic memory refers to executive system that coordinates and integrates their
the visual store. functions.
 II. Short-term memory 11

RESPONSE RESPONSE
OUTPUT GENERATOR

LONG-TERM
REHEARSAL STORE
S BUFFER
E
N
S SHORT-TERM
O STORE
R
STIMULUS Y R
INPUT Self-addressable memory bank
E
G not subject to decay
Memory bank subject
I
to rapid decay
S
T
E
R CONTROL PROCESSES
Stimulus analyzer programs
Alter biases of sensory channels
Activate rehearsal mechanism
Modify information flow from SR to STS
Code and transfer information from STS to LTS
Initiate or modify search of LTS
Heuristic operations on stored information
Set decision criteria
Initiate response generator

Figure 6A The multi-store memory model. Memory can be broadly categorized as having a few basic components, delineated by the
timing of their participation. Sensory input impinges upon sensory organs (eyes, ears, etc.) and is held very briefly as a perception in a
sensory register. Information then traverses to a short-term store where it is held (and potentially rehearsed and processed) for seconds
to minutes. From the short-term store the information may be passed on for long-term storage for minutes to years. Higher-order control
mechanism and processes such as attention-related systems orchestrate the overall process. (SRsensory register; STSshort-term store;
LTSlong-term store). Adapted, with permission, from the work of Shiffrin and Atkinson (1969). Storage and retrieval processes in long-
term memory. Psychological Review 76:179193. Copyright 1969 by the American Psychological Association.

C. The Prefrontal Cortex and Working Memory

Central
executive What brain region does the work in working mem-
ory? There is very strong experimental support that
the prefrontal cortex (PFC) is one anatomical site sub-
serving working memory. The PFC in humans is large,
is located in the rostral part of the frontal lobes, and
Visuospatial Episodic Phonological occupies about one-third of the cerebral cortex. The
sketchpad buffer loop PFC is immediately rostral to the premotor and motor
cortices, and is extensively interconnected with other
parts of the cerebral cortex and the hippocampus.
The laboratory of the late Patricia Goldman-Rakic
Visual Episodic pioneered studies in non-human primates that dem-
Language
semantics LTM
onstrated a role for the PFC in working memory.
These elegant studies combined discrete anatomical
lesioning approaches, pharmacologic studies, and
Fluid systems Crystallized systems
direct recordings in vivo from the PFC during work-
Figure 6B Baddeleys working memory module. An executive ing memory tasks. More recently, these studies have
control system regulates the integration of three basic components been reinforced by studies in humans using functional
of the working memory system. The system overall coordinates the magnetic resonance imaging (fMRI).
processing of visual sensory information, verbal language, and infor- Altogether, a convincing case has been made that
mation recalled from long-term episodic memory stores. (LTMlong-
term memory). Adapted with permission from Baddeley (2001). Is
the PFC contributes to working memory, and indeed
working memory still working? Am. Psychol. 56:849864. Copyright Goldman-Rakic and colleagues have proposed that
2001 by the American Psychological Association. different PFC subregions contribute to different
12 1. Introduction

Prefrontal cortex
Working memory

Response
Ventrolateral PFC Dorsolateral PFC
Non-spatial memory Spatial memory
(color, shape, etc.)

Figure 7 Anatomical subdomains of working memory. This

model is based on work by Goldman-Rakic and co-workers.

Ha Dis Se
bitu hab nsi
atio itua tiza
n tion tion
Short-term memory Long-term memory
Working memory Figure 9 Some simple non-associative forms of learning.
Habituation, dishabituation and sensitization are illustrated. Each
circle represents a hypothetical response to an environmental stim-
ulus. Habituation is a decrease in response (arbitrarily defined in
this schematic example) with repeated presentation of the stimulus.
Action potential firing Persisting molecular and Dishabituation is a recovery to normal baseline response when the
sustained neural circuit cellular changes animal receives a different environmental stimulus. Sensitization
activity changes in synaptic
is an increase in the magnitude of the response above the original
structure
baseline.
anatomical circuit alterations

Figure 8 Mechanisms for storing short-term memory are dis-

tinct from those underlying long-term memory.
III. Unconscious learning

components of the working memory system. A. Simple Forms of Learning

Specifically, they have proposed that persisting neuro- In this section we will explore several simple,
nal activity in the ventrolateral PFC contributes to non- i.e., non-associative, forms of learning. Keep in mind
spatial short-term memory (the color and shape of an that even those forms of learning that exhibit them-
object, for example) while the dorsolateral PFC contrib- selves in a fairly straightforward manner at the
utes to spatial short-term memory (see Figure 7). behavioral level involve elaborate underlying cellu-
lar and molecular machinery. In this section we will
D. Reverberating Circuit Mechanisms Contrast emphasize that several forms of non-associative learn-
with Molecular Storage Mechanisms for ing are exhibited by animals, including: habituation;
dishabituation; and sensitization (see Figure 9). These
Long-Term Memory
forms of learning involve altered responses to a single
The mechanisms underlying short-term memory stimulus, and do not necessitate the animal forming
and working memory involve persistent firing of neu- any association between one environmental stimulus
rons within the PFC and elsewhere in the CNS. Thus, and anotherthat is, these forms of learning are non-
the memory trace for short-term memory is based in a associative. They also can occur unconsciously (see
repetitively firing neural circuit actively encoding and Figure 3), generally requiring neither conscious per-
holding information. It is important to emphasize that ception of environmental stimuli nor conscious recall
this mechanism contrasts with the mechanisms under- of information.
lying long-term memory, which do not rely on persist- Perhaps the simplest form of learning in exist-
ent or reverberating neuronal action potential firing ence is habituation; for example when an animal is
for their persistence (see Figure 8). Thus, short-term repeatedly presented with an innocuous environmen-
memory storage and long-term memory storage man- tal stimulus, the animals response to that stimulus
ifest a fundamental mechanistic difference. Moreover, decreases over time. For example, if someone moves
the fact that long-term memories can be maintained in from a small, quiet town to a street-level apartment
the absence of ongoing action potential firing (at least in Manhattan, typically at first the street noises in
for long periods of time) means that the fundamental the city are disturbing. However, over time, the new-
unit of information storage, the engram, must reside comer becomes accustomed to the new environment,
wholly or in part at the molecular and cell structural and the street noise is no longer so bothersome. This
level in the case of long-term memory. type of phenomenon is referred to as habituation. The
 III.Unconscious learning 13

Box 3

Aplysia in its natural habitat

Given the popularity of Aplysia as an experimen-

tal system, one might be tempted to think of Aplysia as
being indigenous to the aquaria of neurobiology labo-
ratories. However, the most widely studied Aplysia spe-
cies, californica, lives in the cool Pacific waters off the
California coast. Aplysia spends its time in the tidal and
near-coastal zones, where it feeds on a diet of seaweed.
Except for the buffeting of the ocean waves and currents
(and, one must assume, the occasional curious scuba
diver), Aplysia lead a fairly peaceful existence. They are
unsavory to fish and have very few natural predators;
however, Aplysia can serve as prey to certain types of
sea anemones. When an Aplysia is seriously perturbed,
it exhibits its most dramatic behavioral response; inking.
Aplysia possess an ink gland and can release a cloud of
viscous purple ink, similar to the well known octopus.
Although the precise function of the inking is unknown,
two popular ideas are that the ink may contain noxious
compounds to help ward off predators, or may serve
to camouflage the animal from potential attackers. A
strong aversive stimulus such as one that elicits inking
by Aplysia also results in sensitization of the animal. For
some period of time after inking an animal will exhibit
enhancement of its baseline defensive withdrawal
responses. This ethologically relevant form of behavior
modification is the basis for laboratory study of sensiti-
zation in Aplysia.

teleologic explanation for habituation is that over time at their doorstep, street noises may once again become
animals learn to ignore environmental stimuli that noticeable. It is worth noting that dishabituation is
carry no unique informational content. a useful tool to distinguish habituation from fatigue.
Habituation is a very robust behavioral phenom- A habituated response can be overcome by a dishabit-
enon that exhibits itself in many formsessentially all uating stimulus; however, a decreased response due
baseline behavioral responses more complex than the to fatigue cannot.
purest reflex responses habituate. Some of the more Animals can also learn to become hyper-responsive
well studied habituation phenomena experimentally to an environmental stimulus, a phenomenon known
are habituation of the Aplysia californica gill-and- as sensitization. Sensitization is defined as an increased
siphon defensive withdrawal response and habitua- response over and above the normal baseline response
tion of reflexive leg-lifting in Drosophila. Habituation which occurs in response to an environmental signal.
is also frequently encountered outside the labora- Sensitizing stimuli typically can elicit an augmentation
tory setting; in particular it is frequently observed by in response from either a non-habituated or a habitu-
teachers in the classroom lecture environment. ated starting point. In the second scenario a component
After a response is habituated, if you present of the increased responsiveness must, by definition,
another, unique, stimulus, dishabituation can occur. then be described as dishabituation (see Figure 9).
For example, even after becoming habituated to street Keep in mind that in some ways the definitions
noises, if one is expecting a visitor to be dropped off of habituation, dishabituation, and sensitization are
14 1. Introduction

Box 4

Hermissendathe good-looking one in the family

While even a dedicated neurobiologist would be

hard-pressed to describe Aplysia as aesthetically attrac-
tive, another popular invertebrate species used in stud-
ies of learning and memory is a clear winner in any
molluscan beauty contest. With its bright coloration
and striking profile, Hermissenda is the closest thing to
a poster child available among the invertebrate species
commonly studied by neurobiologists. Hermissenda is
not just all looks and no brains, however. This system
has been used to study the cellular and molecular basis
of a particular form of associative learning exhibited by
the animal. Hermissenda are normally phototactic; that
is, they will move toward a lighted area. However, if the
animal is trained that light predicts an upcoming aver-
sive stimulus, in this case turbulence in the water sur-
rounding the animal, the normal phototactic response
is suppressed. The laboratories of Dan Alkon and Terry
Crow have been instrumental in discovering the neuro-
nal circuitry, cellular physiology, and molecular mecha-
nisms underlying this form of associative conditioning.

arbitrary. In the natural setting animals are constantly on easily studied, simple forms of learning in special
modifying their behaviors in response to the ongoing preparations that lent themselves to experimental
barrage of environmental signals. Thus, it is difficult investigation at the cellular level. In particular, the
to determine what a baseline response is outside of work of Eric Kandel (Figure 10) and his colleagues
a stringently controlled experimental setting. allowed enormous progress in our understanding of
All of these non-associative forms of learning can the cellular basis of behavior in general, and learning
exhibit themselves in either short-term or long-term and memory specifically. Kandel and his colleagues
forms. The duration of the memory for a learned event Tom Carew, Jack Byrne, and Bob Hawkins, along with
depends on the number of times an animal experi- many others, have used the simple marine mollusk
ences a behavior-modifying stimulus. For example, Aplysia californica (Figure 11) to great effect to study
a single sensitizing stimulation may elicit sensitiza- the behavioral attributes and cellular and molecular
tion that lasts only a few minutes, whereas repeated mechanisms of learning and memory.
stimulations will likely result in sensitization lasting
hours to days (see Figure 4, for example). Repeated
presentations of multiple training trials can elicit sen- B. Unconscious Learning and Unconscious Recall
sitization lasting for weeks.
Motor Learning
One exciting area of contemporary neurobiologi-
cal research is to try to understand the basis for short- Motor learning, skills, and habits are the classic exam-
term and long-term non-associative learning. Starting ples of unconsciously learned and unconsciously recalled
in the 1960s the mechanisms underlying habituation memories. Walking is a good example. Walking is an
and sensitization began to be worked out at the cel- extremely complex task involving intricate motor move-
lular and biochemical level. Part of this watershed ments, which we generally perform automatically and
of new understanding of the basis of learning and mem- with great facility. We learned to walk unconsciously as
ory came about as a result of the insight to capitalize small children and, if anything, trying to exert conscious
 III.Unconscious learning 15

contractions and hand movements are taking place com-

pletely below the level of conscious thought.
While complex unconscious processes go into the
initial establishment of learned motor patterns, in
some cases such as speech and walking, there is prob-
ably also a complicated interaction of developmental
processes with signals generated in response to envi-
ronmental stimuli. As mentioned above, in the early
stages of many types of motor learning there is con-
scious involvement, the need for which disappears
over time as part of the learning process. The circuitry
and cellular mechanisms underlying motor learning
are quite complex, involving the motor cortex, basal
ganglia including the neostriatum, and cerebellum.
The site of memory storage for most types of motor
memory involve or have access to the principal cir-
Figure 10 Dr Eric Kandel. Dr Kandel is a University Professor cuits which mediate the behavioral motor pattern,
at Columbia University and Nobel Prize-winner who led pioneer- such as the motor cortex, basal ganglia, and spinal
ing studies on the cellular basis of learning and memory. cord motor neurons. A discussion of these systems
is presented in Chapter 2 as part of the discussion of
human memory systems.
Some motor memories are subject to limited con-
scious recall, but in most cases trying to replay a
motor memory with too much conscious control sim-
ply messes things up. This is likely a component of the
common choking component of sports, although
stress-induced release of modulatory neurotransmit-
ters which affect performance is also certainly a factor.
It is interesting that the unconscious aspect of motor
recall has made it into popular sports lingo. When an
athlete is at the top of his or her game they are typi-
cally referred to as being unconscious.

C. Unconscious Learning and Subject

Figure 11 Aplysia Californica. Aplysia, a nudibranch mollusk
to Conscious Recall
found in the cool waters off the coast of California, popularized for
its use in studies of simple forms of learning and memory. The forms of learning we have talked about so far
are non-associative. In habituation, sensitization, etc.,
nothing is learned about the relationship or associa-
control over our walking as adults likely leads to an tion of one event with another. We next move on to
awkward gait. a more complex form of learning where a predictive
Another example of unconscious learning is learning relationship is learnedan animal learns that one
to play an instrument such as the guitar or piano, at least environmental stimulus reliably predicts another.
as concerns the motor components. Repetition allows An important set of nomenclature in this area arose
the development of finely tuned motor patterns that can out of the pioneering work of Ivan Pavlov (Figure 12).
be recalled without conscious thought. Learning of the Pavlov and his co-workers studied associative con-
motor components also occurs without much conscious ditioning of the salivary response of dogsstudies
control, although certainly there is conscious involve- indeed so classic that the terms classical condition-
ment when the initial motor patterns are beginning to ing and Pavlovian conditioning are now used syn-
be laid down. Even in this case, though, one does not onymously with associative conditioning. Pavlov
consciously work out the pattern of firing of individual knew, as does anyone that has ever owned a dog,
musclesindeed we by-and-large dont have very much that when a dog is presented with a food stimulus a
control over the contraction of single muscles and are not strong salivatory response is elicited (see Figure 13).
really conscious of them as single units. When we learn This is a natural response, of course, and this saliva-
to play an instrument, a multitude of complex muscle tion is referred to as the unconditioned response, and
16 1. Introduction

correspondingly the food stimulus is referred to as the example the reliability of a tone for predicting a sub-
unconditioned stimulus. Pavlovs breakthrough reali- sequent food presentation. This type of learning is
zation, which he subsequently rigorously documented profoundly important for survival in any natural envi-
and studied, was that he could train dogs to associate ronment, and for this reason has been robustly selected
a neutral stimulus, such as the ringing of a bell, with for in animal evolution. Stated another way, associa-
the food stimulus. Over time the dog would form an tive learning allows the neural encoding of cause-and-
association between the bell and the food, and Pavlov effect relationships. The stable formation of a memory
found that the bell alone would ultimately cause a trace, such that an accurate record of cause-and-effect
salivatory response just like the food did. The bell- relationships is available for future reference, provides
elicited salivation was termed the conditioned res- such a pronounced competitive advantage that this
ponse, and correspondingly the bell tone was termed form of learning is typically quite vigorous.
the conditioned stimulus (Figure 13). The importance of this last point cannot be over-
In associative learning an animal learns the pre- stated! Nature has selected for a robust capacity of
dictive value of one stimulus for another, in Pavlovs nervous systems to accurately reflect one of the prin-
cipal physical laws that govern the real world: cause
and effect. Thus, nervous systems of all sorts have
evolved to the best of their capacity sophisticated and
robust circuit, cellular, and molecular mechanisms to
encode these types of information.
Generally associative learning is quite reliable
obviously the accuracy of storing cause-and-effect rela-
tionships is of paramount importance and has been
selected for evolutionarily. This is one significant fac-
tor in the popularity of studying associative learning
experimentallythe learned behaviors are observable,
relatively rapidly acquired, and reliably expressed.
However, this is not to say that associative learning
is flawless. Numerous examples exist in the literature
and anecdotally of animals having mislearned associ-
ations. For example, one of my colleagues who works
Figure 12 Ivan Pavlov pioneered the study of associative con- with Macaque monkeys had a monkey who learned
ditioning, studying modification of reflex responses in dogs. that certain visual stimuli predicted the subsequent

Conditioned
stimulus

Unconditioned
stimulus
Unconditioned
response

Conditioned
stimulus

Conditioned
response
Figure 13 Pavlovian associative conditioning of the canine salivary response. Repeated pairings of an auditory cue with food causes the
animal to learn the predictive value of one for the other. See text for details.
 III.Unconscious learning 17

arrival of a food reward. However, the animal also Delay conditioning US

learned that it was necessary to wave his hand in CS
an idiosyncratic way in order for the food reward to
be delivered. Of course, in reality the hand movement
was entirely superfluous to the task and the reward Time
delivery. B. F. Skinner recorded several similar cir-
cumstances in training pigeons in associative learning
tasks, where in some cases elaborate but unneces- Trace conditioning
US
sary motor patterns were executed before the animal CS
pecked an object to receive a food reward. Skinner
termed these behaviors superstitious behaviors,
a somewhat loaded term that is anthropomorphic, Time
but not without appeal. Regardless of the terminol- Figure 14 Delay and trace conditioning. Associative condition-
ogy, it is clear that these are examples of associative ing falls into two broad categoriesdelay conditioning, and trace
learning gone awry. Presumably what has happened conditioning. In trace conditioning an intervening time interval is
is that early on in the training, the animal has errone- introduced between the termination of the CS and the onset of the
US. Trace conditioning involves the hippocampus.
ously associated some movement on their part with
the food reward, and formed a lasting but inaccurate
memory that executing the movement is necessary
to receive the reward. I bring up these examples as simultaneously with a CS after the CS has been pre-
indications of the robustness of associative mem- sented continuously for some delay period.
ory, but with the interesting twist that as with all Trace conditioning refers to a conditioning proto-
robust systems there is an attendant possibility of col where the CS is presented, terminated, and fol-
error-proneness. lowed after some intervening period by the US. The
Against this backdrop it is then interesting to CS and US never overlap in time, and are temporally
consider that associative learning depends on two contiguous in the sense that they are presented closely
attributes of the environmental stimulicontigu- in time, but never simultaneously. The term trace
ity and contingency. Contiguity refers to the property arises from the fact that some memory trace for the
of the stimuli occurring coincidentally, that is over- CS must be preserved over time so that it can subse-
lapping in time or one immediately after the other in quently be associated with the US.
time. This captures Natures rule of cause and effect. The distinct use of the two terms delay condi-
Environmental stimuli are generally perceived simulta- tioning and trace conditioning may seem like
neously with or immediately after the events that cause scientific hyper-semanticism, because the two proto-
them. Contingency refers to the ordering of the stim- cols seem so similar. However, the reasonably subtle
ulithat one stimulus consistently precedes the other alteration of introducing a brief intervening time span
in onset. This captures the predictive value of one event between CS and US brings entirely new neuronal cir-
for the other; that is, in nature the cause will always cuits to bear on the cognitive processing involved.
precede the effect. In the examples of mislearning in the Indeed, trace conditioning requires the hippocampus,
previous paragraph the animals presumably misrepre- whereas delay conditioning does not. This hippo-
sented contiguous stimuli as also being contingent. campus-dependence of trace associative learning has
The issue of contiguity in associative conditioning been largely studied in rodents, but elegant studies
raises the consideration of two basic types of classi- in Larry Squires laboratory have demonstrated that
cal associative conditioning: delay conditioning and humans with hippocampal lesions also have defi-
trace conditioning (see Figure 14). The type of classi- cits in trace associative conditioning (7). Thus, delay
cal associative conditioning we have discussed so far and trace conditioning differ fundamentally in their
is referred to as delay conditioning. This term derives underlying anatomy and relevant circuitry, so much
from the typical timing of this type of associative con- so that they indeed are quite different forms of learn-
ditioning protocol experimentally. For example, if one ing. It is for this reason that making the reasonably
is training an animal to learn that a tone predicts a small move from delay conditioning to trace condi-
food reward using a delay conditioning protocol, the tioning progresses us from one category of learning to
tone is started and maintained continuously until the another entirely.
food reward is presented. Thus, the onset of the CS is The recall of trace conditioning has mapped onto
followed by a delay before the onset of the US. With it a temporal component as well. Re-experiencing the
delay conditioning the CS and US are contiguous and CS after trace conditioning has occurred allows for
overlappingin other words the animal receives a US conscious recollection of the US, during the trace
18 1. Introduction

period. For example, lets say that I am trained that One specific example of fear conditioning involves
a tone preceeds a foot shock by five seconds. During the delivery of an innocuous acoustic cue (CS) paired
testing, when I hear the tone I have five seconds dur- with a mild foot shock (US) within a novel environment
ing which I am expecting the foot shock to be deliv- (see Figure 15). When tested 24 hours after training,
ered. Because of this aspect of the possibility of rats, mice, and other rodents exhibit marked fear,
conscious anticipation, trace conditioning is our first measured by freezing behavior or other reflex fear
example of learning that can occur unconsciously but responses, in response to representation of either the
can be subject to conscious recall (see Figure 3). context (contextual fear conditioning) or the auditory CS
delivered in a different context (cued fear conditioning).
Both cued and contextual fear conditioning have been
D. Operant Conditioning shown to be dependent upon the amygdala, whereas
contextual fear conditioning also involves the hippo-
Pavlovs dogs were passive participants in their
campus. We will return to these two forms of learning
learning experience. They did not have to do any-
in Chapter 4, where we discuss rodent behavioral mod-
thing beyond perceiving the environmental stimulus,
els of learning in more detail.
after which natural reflexes took over and an uncon-
Conditioned taste aversion is another form of
scious salivatory response occurred. This type of
associative learning; in this case, an animal learns to
learning is distinct from learning paradigms where
associate the novel taste of a new foodstuff (CS) with
a voluntary motor response is elicited. Conditioning
subsequent illness (US) resulting from ingestion of
where the animal is required to execute a voluntary
some toxic agent (see Figure 16). The adaptiveness
motor response is referred to as operant conditioning.
of this form of learning should be apparent; by pre-
It is important to bear in mind that the distinction is
venting subsequent ingestion of poisonous foods sur-
a practical one, based in experimentation. Operant
vival is greatly enhanced. This is obviously a form of
conditioning simply refers to the fact that the experi-
learning that is not very forgiving of multiple train-
menter is quantitating a voluntary movement (not a
ing trials; not surprisingly, animals learn after a single
reflex) as the behavioral output indicating that learn-
pairing of novel taste and toxin to avoid that taste in
ing has occurred. The examples I used above where
future encounters.
monkeys or pigeons were required to push a lever or
One interesting aspect of conditioned taste aver-
peck a button are examples of operant conditioning.
sion learning is the long CS-US interval. Unlike other
Over the years there has been debate over whether
operant conditioning will use different mechanisms associative conditioning paradigms, such as fear con-
from classical associative conditioning, and whether ditioning or eye-blink conditioning where the CS-
operant and classical conditioning should really be US interval is typically on the order of seconds, with
considered as distinct categories. We dont have the conditioned taste aversion the system can tolerate
final answer to this question, but suffice it to say that delays of hours between the CS taste and US toxin.
it appears that operant conditioning will not require This suggests that there are cellular and biochemical
unique cellular or molecular mechanismslikely the events initiated by the taste stimulus alone that are
mechanistic differences will be confined to the types likely to be long-lasting. Indeed, novel tastes alone
of neuronal circuitry involved. trigger memory formation automatically, as can be
measured by increased food consumption upon repre-
sentation of a foodstuff (Figure 16). This phenomenon
E. Currently Popular Associative Learning is referred to as attenuation of neophobia (11).
Finally, comparing and contrasting fear condition-
Paradigms
ing with conditioned taste aversion raises a final gen-
Two associative learning paradigms that are used eral attribute of associative conditioning, referred to
extensively in the modern laboratory for studying as salience. We discussed above that contingency and
learning are conditioned fear (89) and conditioned contiguity are two hallmarks of associative condition-
taste aversion (10). In both paradigms, animals learn ing, and salience is a term used to refer to the third
an association between a neutral conditioned stimulus general property. Salience refers to the fact that ani-
and an aversive unconditioned stimulus. Both serve as mals do not in general learn to associate conditioned
powerful examples of classical Pavlovian condition- stimuli and unconditioned stimuli that are not typi-
ing, and both result in robust, long-lasting memory cally paired in their natural environment. For exam-
after even a single CS-US pairing. These two behavio- ple, nausea-inducing stimuli are by far much more
ral paradigms are also accommodating to researchers robust producers of taste aversion than are generic
because the neuroanatomical pathways underlying painful stimuli that may robustly support other types
the learning are fairly well-established. of aversive conditioning. Similarly, pairing nausea
 III.Unconscious learning 19

TRAINING

Animal is placed in novel context

Hears a tone
Receives foot shock

CONTEXTUAL TEST CUED TEST

Animal is returned to same context Animal is placed in modified context

Test for freezing behavior Hears a tone
Test for freezing behavior
Figure 15 Fear conditioning. Fear conditioning is a form of associative conditioning in which an aversive, fear-evoking stimulus is
paired with a novel environmental cue. A wide variety of environmental stimuli can be used for fear conditioning, such as places (contexts),
auditory cues, visual cues, odors, etc.

Behavioral procedures used to assess novel taste learning

NEOPHOBIA TASTE AVERSION

?
DAY 1 10

DAY 1 10
LiCl

DAY 2 10 DAY 2 10

Measure increased consumption Measure decreased consumption

as index of long-term memory as index of long-term memory
of novel taste of novel taste

Figure 16 Taste learning. Taste learning is a robust and automatic form of learning in animals. Two types of assessments generally used
to evaluate taste learning are attenuation of neophobia, in which an animal learns that a taste is not dangerous, and conditioned taste aver-
sion, in which an animal learns that a given taste is dangerous. These can be measured experimentally by giving a single exposure to a novel
food on Day 1 (ten minutes of Nutri-grain bar in this case) and monitoring the animals response to representation of that same stimulus 24
hours later (Day 2). If the animal finds the new food to be non-aversive, consumption of the food will be increasedattenuation of neopho-
bia. If an aversive stimulus is paired with the novel food (e.g., a lithium chloride injection), then the animal will exhibit a decreased con-
sumption of the food on Day 2conditioned taste aversion. See Chapter 4 for additional details.

with visual stimuli or auditory cues is not very effec- and a term that is used to describe this is that the
tive at aversive conditioning to these stimuli. Clearly, stimulus is salient to the animal, in other words, the
evolution has operated to select for robust learning of stimulus is likely to be pertinent to the animal under
associations that can occur in the natural environment, the given condition.
20 1. Introduction

Thus, in general, it is not the case that unconscious this book will also depend on a large number of other
associative learning operates such that any two envi- factors, such as motivation, attention, level of arousal,
ronmental signals can be associated. This likely arises etc. Thus, human declarative learning, and likely most
from a combination of factors. First, there is a degree analogous forms of learning in animals, is subject to a
of anatomical specialization in the CNS such that par- wide variety of modulatory factors.
ticular functions are parsed out into particular areas. For example, particularly robust memory for single
Thus, as a practical matter, the central processing of events is typically referred to as flashbulb memory
two environmental stimuli may never touch each in humans. There are several examples of this type
other in the brain, and therefore can never be asso- of memory that many Americans have shared, the
ciated. In this instance the stimuli touching each most recent example being the terrorist destruction
other can be taken quite literally in that some ana- of the Twin Towers in New York City. Like most peo-
tomical cross-connection must be made. Conversely, ple, I remember vividly how I learned of the attacks,
for any association to take place the underlying neu- and I am sure I will never forget seeing live on televi-
ral circuits processing the environmental informa- sion the second tower collapse. Flashbulb memories
tion must be able to connect anatomically. It is only are usually associated with a high state of arousal
possible to draw an associative learning circuit if the or a high level of emotional valencean example
two stimuli being paired impinge upon each other at of the strong modulatory influences that learning is
some point. subject to.
This probably seems like a statement of something As are the other types of conscious learning we will
that is intuitively obvious. However, if associative discuss in this section, declarative learning is depend-
learning requires that two information-processing cir- ent on the hippocampusin later sections of the book
cuits connect with each other, this must of necessity we will return to the importance of modulatory influ-
utilize molecular and chemical processes. A descrip- ences on hippocampus-dependent learning, and dis-
tion of these types of processes, in particular molecular cuss some likely molecular mechanisms underlying
mechanisms that can contribute to associative events, this effect.
is a central theme of later chapters of this book. It is difficult to model declarative learning in non-
Overall, we have seen in this section that various human animals, because the behavioral output for
models of associative conditioning have transitioned these types of memories is actually quite subtle and in
us from unconscious processes to conscious processes. most cases it is not even clear what are relevant type
Many associations can be learned unconsciously and of learning might be in lower animals (see Box 5).
expressed unconsciously. However, various types of Partly for this reason most of what we know about
associative learning also begin to recruit conscious declarative learning comes from human studies,
processes. While learned unconsciously they can be in particular studies of patients with hippocampal
recalled consciously. As a generalization, the transi- lesions. These studies will be described in more detail
tion involves recruitment of the hippocampus into the in Chapter 2 (see also references 1213). Suffice it to
learning process. In the following section we will tran- say for our purposes that a number of classic studies
sition to even more complex forms of learning that of humans with hippocampal lesions led to the dis-
also depend on the hippocampus. sociation of declarative from non-declarative forms of
memory in humans.
Declarative memory is that type of memory that is
IV. Conscious learningsubject lost when a human suffers hippocampal damage
to conscious and unconscious this includes the capacity to form memories for facts,
names, places, and personal experiences (Figure 3).
recall
Hippocampal damage results in anterograde amnesia for
these types of memoriesthat is, there is a loss of the
A. Declarative Learning
capacity to form new memories. Old memories (more
Human declarative learning is what we typically than about one year) are largely spared, i.e., there is rel-
think of when we think of learning. This is the con- atively little retrograde amnesia. Non-declarative forms
scious acquisition of new facts, or the formation of of memory such as sensitization, motor learning, delay
memories for events that occur in our lives, which are classical conditioning, etc., are spared in humans with
available for subsequent recall at will. The extent to hippocampal lesions.
which you remember what you read in this book will A final comment on declarative learning is that it is
depend upon the processes of conscious declarative generally associative, although not in the sense of clas-
learning. Of course, the extent to which you remember sical associative conditioning where a cause-and-effect
 V. Summary 21

Box 5

A rodent model of declarative memory?

It is difficult to imagine a rodent model for declara- This is readily demonstrated in the laboratory: when rats
tive memory, but there is one potentially parallel type of or mice are presented with highly palatable solutions
learning in rodents that I will mention briefly. Because of novel tastes, such as saccharin or sucrose, they will
toxic plants and other poisonous foodstuffs coexist with consume small amounts on the first exposure; on subse-
most animals, as described above conditioned taste aver- quent exposures, the animals consume more (see Figure
sion evolved to protect animals from being poisoned out 16). This attenuation of neophobia is a behavioral meas-
of the gene pool. However, avoidance of something that ure of memory for the novel taste, and is part of a proc-
is toxic is not possible if it has been ingested in lethal ess of familiarization to the formerly novel taste. There
quantities, so a supplementary behavior has also evolved is a fairly clear consensus that the insular cortex is the
to protect animals from toxic foods. Neophobia is the char- primary site of learning and memory for novel tastes, so
acteristic fear of novel foods, and ensures that animals this form of learning is clearly not strictly analogous to
ingest only small quantities, as if to sample the food to human declarative learning. However, it does depend
determine if it is safe to eat. If the animal develops illness, on the cerebral cortex as its storage site, as is likely in
a conditioned taste aversion results, and this foodstuff human declarative memory. Furthermore, it is reason-
will be avoided on future encounters. If no illness results ably analogous to a human learning a fact, in this case
and assuming the food is reasonably palatable, animals what something tastes like, and having that information
will increase their intake on subsequent exposures. available for conscious recall.

relationship is learned. Most declarative learning B. Spatial Learning

does not take place in a cognitive vacuum, but items
A final example of hippocampus-dependent learn-
are typically learned in the context of other related
ing in both humans and lower animals is spatial
facts or objects. A good example of this for illustra-
learning. Obviously animals must learn to navigate
tive purposes is learning someones name. Learning
their environment, and learn to associate particular
a name is certainly a declarative learning event, and
places with particular items or events. This type of
you can list off the names of all the people you know
learning has been the classically defined learning sys-
well as a reiteration of a list of facts. However, each
tem in which involves the hippocampus. A wide vari-
name also serves as a descriptor of an individual and
ety of different studies have shown that molecular or
is associated with that person, their face, their house,
anatomical lesions of the hippocampus lead to spa-
etc. This type of multiple association for learned
tial learning deficits, in both humans and lower ani-
facts (i.e., declarative learning) is the rule rather than
mals. Also, direct measurements of a wide variety of
the exception. It is likely that most declarative learn-
molecular and physiologic changes have been shown
ing occurs as learning something within a variety of
to correlate with spatial learning. These topics will be
contexts: other facts or places with which the fact is
discussed in greater detail in the next chapter.
associated. This point is important to keep in mind as
we begin to explore the molecular basis for declara-
tive learning. It is certainly possible that many of
the molecular mechanisms that are discovered as V. Summary
subserving what we have defined as associative con-
ditioning may translate directly as mechanisms This chapter has described a number of basic
contributing to declarative learning. Stated more attributes of learning and memory, based largely on
strongly, at this point it is appropriate to hypoth- the psychological study of these phenomena. We also
esize that associative molecular mechanisms will be have introduced and discussed a number of terms
part of the molecular infrastructure of declarative related to memory and its study. Mastery of these
learning. basic terms and an understanding of their utilization
22 1. Introduction

will be critical for your comprehension of the remain- Squire, L. R. (2004). Memory systems of the brain: a brief history
der of this textbook. and current perspective. Neurobiol. Learn. Mem. 82(3):171177.
Squire, L. R., and Kandel, E. R. (2008). Memory: From Mind to
There are a number of basic take-home messages Molecules, 2nd ed. New York: Roberts and Company.
for this chapter. Tronson, N. C., and Taylor, J. R. (2007). Molecular mechanisms
of memory reconsolidation. Nature Reviews Neuroscience
1. Memory is not monolithicthere are many 8:262275.
different types of learning and memory, most of
which are subserved by different anatomical areas
of the nervous system. This principle is distilled in Journal Club Articles
the multiple memory systems concept. Bechara, A., Tranel, D., Damasio, H., Adolphs, R., Rockland, C., and
2. Almost all forms of memory have both short-term Damasio, A. R. (1995). Double dissociation of conditioning and
and long-term forms. declarative knowledge relative to the amygdala and hippocam-
3. The mechanisms underlying short-term memory pus in humans. Science 269(5227):11151118.
Knowlton, B. J., Mangels, J. A., and Squire, L. R. (1996). A neo-
are distinct from those underlying long-term
striatal habit learning system in humans. Science 273(5280):
memory. In general, brief forms of short- 13991402.
term memory are sustained by repetitive or Packard, M. G., and McGaugh, J. L. (1996). Inactivation of hippoc-
reverberating action potential firing, while long- ampus or caudate nucleus with lidocaine differentially affects
term forms of memory are sustained by persisting expression of place and response learning. Neurobiol. Learn.
Mem. 65(1):6572.
molecular and cellular modifications.
For more informationrelevant topic chapters from: John H. Byrne
4. Memory can be broadly subdivided into non- (Editor-in-Chief) (2008). Learning and Memory: A Comprehensive
declarative and declarative forms, and similarly Reference. Oxford: Academic Press (ISBN 978-0-12-370509-9).
into unconscious and conscious forms. (1.02 Roediger, H. L. III, Zaromb, F. M., and Goode, M. K. A
5. Both unconscious and conscious forms of Typology of Memory Terms. pp. 1124; 1.03 Capaldi, E. J., and
Martins, A. History of Behavioral Learning Theories. pp. 2539; 1.04
learning can be manifest in many different
Nadel, L. Multiple Memory Systems: A New View. pp. 4152; 3.01
ways, for example associative and non-associative Eichenbaum, H. Introduction and Overview. pp. 18; 3.02 White,
forms, operant forms, etc. These subtypes N. M. Multiple Memory Systems in the Brain: Cooperation and
vary depending on the particular sensory Competition. pp. 946.)
inputs that trigger the learning, and depending
on the behavioral output that manifests the
memory. References
1. Squire, L. R., and Kandel, E. R. (1999). Memory: From Mind to
Molecules (distributed by W. H. Freeman and Co.). New York:
Scientific American Library.
Further Reading 2. Eichenbaum, H. (2001). The hippocampus and declarative
Baddeley, A. (2001). Is working memory still working? Am. memory: cognitive mechanisms and neural codes. Behav. Brain
Psychol. 56:849864. Res. 127:199207.
Eichenbaum, H. (2002). The Cognitive Neuroscience of Memory. New 3. Kandel, E. R., and Squire, L. R. (2000). Neuroscience: breaking
York: Oxford University Press. down scientific barriers to the study of brain and mind. Science
Eichenbaum, H., Yonelinas, A. P., and Ranganath, C. (2007). The 290:11131120.
medial temporal lobe and recognition memory. Annu. Rev. 4. Eichenbaum, H., and Cohen, N. J. (2001). From Conditioning to
Neurosci. 30:123152. Conscious Recollection: Memory Systems of the Brain. Upper Saddle
Gold, P. E. (2004). Coordination of multiple memory systems. River, NJ: Oxford University Press.
Neurobiol. Learn. Mem. 82(3):230242. 5. Vianna, M. R., Szapiro, G., McGaugh, J. L., Medina, J. H., and
LeDoux, J. E. (2001). Synaptic Self: How Our Brains Become Who We Izquierdo, I. (2001). Retrieval of memory for fear-motivated
Are. New York: Viking. training initiates extinction requiring protein synthesis in the
McDonald, R. J., Devan, B. D., and Hong, N. S. (2004). Multiple rat hippocampus. Proc. Natl Acad. Sci. USA 98:1225112254.
memory systems: the power of interactions. Neurobiol. Learn. 6. Nader, K., Schafe, G. E., and LeDoux, J. E. (2000). The labile
Mem. 82(3):333346. nature of consolidation theory. Nat. Rev. Neurosci. 1:216219.
McDonald, R. J., and White, N. M. (1993). A triple dissociation of 7. Clark, R. E., and Squire, L. R. (1998). Classical conditioning
memory systems: hippocampus, amygdala, and dorsal stria- and brain systems: the role of awareness. Science 280:7781.
tum. Behav. Neurosci. 107(1):322. 8. LeDoux, J. E. (2001). Synaptic Self: How Our Brains Become Who
McNaughton, B. L., Battaglia, F. P., Jensen, O., Moser, E. I., and We Are. New York: Viking.
Moser, M. B. (2006). Path integration and the neural basis of the 9. Quirk, G. J., Repa, C., and LeDoux, J. E. (1995). Fear conditioning
cognitive map. Nat. Rev. Neurosci. 7(8):663678. enhances short-latency auditory responses of lateral amygdala
Milner, B., Squire, L. R., and Kandel, E. R. (1998). Cognitive neuro- neurons: parallel recordings in the freely behaving rat. Neuron
science and the study of memory. Neuron 20(3):445468. 15:10291039.
Shiffrin, R. M., and Atkinson, R. C. (1969). Storage and retrieval 10. Berman, D. E., and Dudai, Y. (2001). Memory extinction, learn-
processes in long-term memory. Psychological Review ing anew, and learning the new: dissociations in the molecular
76:179193. machinery of learning in cortex. Science 291:24172419.
 References 23

11. Swank, M. W., and Sweatt, J. D. (2001). Increased histone 13. Milner, B., Squire, L. R., and Kandel, E. R. (1998). Cognitive
acetyltransferase and lysine acetyltransferase activity and neuroscience and the study of memory. Neuron 20:445468.
biphasic activation of the ERK/RSK cascade in insular cortex 14. Kandel, E. R. (2001). The molecular biology of memory
during novel taste learning. J. Neurosci. 21:33833391. storage: a dialogue between genes and synapses. Science 294:
12. Eichenbaum, H. (1999). The hippocampus and mechanisms of 10301038.
declarative memory. Behav. Brain Res. 103:123133.