You are on page 1of 53

Cognitive Neuroscience

‘The Brain Story’

Vaia Lestou
A brief History of Cognitive Neuroscience

• Ancient humans although they wondered extensively about

the nature of human feelings, memories, attention,
communication, motion and many other ‘cognitive
functions’, they had one big problem.

• They did not have the ability to systematically explore the
mind through experimentation.
A brief History of Cognitive Neuroscience

• But if you can observe, manipulate & measure then you

can start to determine how the brain gets its job done
Debrück (1986) ‘Mind From Matter?’

• If you want to understand how a biological system works

then a laboratory is needed and experiments are essential.
A brief History of Cognitive Neuroscience

• Enigma of whether the brain works in concert or

parts of the brain work independently is still the
focus of contemporary research
– face area
– specialised only for faces?
– or objects as well?
A brief History of Cognitive Neuroscience

• Franz Joseph Gall & J.G.

Spurzheim claimed that the brain
was organised into 35 specific
– founders of phrenology in the
early 19th century
• Functions ranged from language
and colour perception to hope and
• If a person used one of the
faculties more than the others the
brain representation area grew
(bump in the skull idea!)
A brief History of Cognitive Neuroscience

• Gall and colleagues

believed that by studying
carefully the skull of a
person you could go a
long way in describing the
personality of the person
inside the skull
• Anatomical Personology
A brief History of Cognitive Neuroscience

• P.J.M. Flourens (1794-1867) challenged Gall’s

localisation views
– bird experiments
• According to Flourens(1824) :‘All sensations, all
perceptions and all volitions occupy the same seat
in these (cerebral) organs. The faculty of
sensation, percept and volition is then essentially
one faculty.’
A brief History of Cognitive Neuroscience

• In France Paul Broca

treated a man who had
suffered from stroke
• the patient could
understand language but
could not speak
• the patient’s left frontal
lobe was damaged
• Broca’s area 3D MRI of human brain with Broca's area highlighted in red
A brief History of Cognitive Neuroscience

• The German Neuroloist

Carl Wernicke in 1876
reported a stroke victim
who could talk freely but
what he said made little
• Patient could not
understand spoken or
written language 3D MRI of human brain with Wernicke's area highlighted in blue

• Wernicke’s area
A brief History of Cognitive Neuroscience

• The most famous of all

physiologists was
Brodmann who analysed the
cellular organisation of the
cortex and characterised
fifty two distinct regions
• It was soon discovered that
the cytoarchitectonically
described brain areas
represent distinct brain
A brief History of Cognitive Neuroscience

• The revolution in our understanding of the nervous system

was brought by Camillo Golgi (Italy) and Ramon y Cajal
• Golgi developed a stain that impregnated individual
• Cajal found that neurons are discrete entities
• He was also the first to suggest that neurons transmit
electrical information in only one directions from the
dendrites to the axonal tip
A brief History of Cognitive Neuroscience

A brief History of Cognitive Neuroscience

A brief History of Cognitive Neuroscience

• The term Cognitive Neuroscience was first coined in a taxi

in the 70s and by that time a new mission was clearly
• neuroscientists were discovering how the cerebral cortex was
organised and functioned in response to simple stimuli
• specific mechanisms were described, such as those relating to
visual perception by Hubel & Wiesel
• models were build to describe how single cells interact to
produce percepts
• and psychologists started to abandon the ideas of learning and
associationism and believed that the behaviours they were
interested in had biological origin and instantiation.
Mission Statement of Cognitive Neuroscience:

‘How the brain actually does enable the mind’

Anatomy of the Brain
• Brain = Cerebral Cortex
• Has two symmetrical Hemispheres
• Each hemisphere consists of large
sheets of layered neurons
• The human cortex is highly folded to
pack more cortical surface into the
• The surface area of the average
human cerebral cortex is about 2200
to 2400cm2
• The infolding of the cortical sheets
are the sulci and the crowns of the
folded tissue are called the gyri
Anatomy of the Brain

• The cortex has a high

density of cell bodies, the
‘grey matter’
• The underlying region is
composed primarily by
axons of neurons and is
known as the ‘white
matter’, they connect the
neurons of the cerebral
cortex to other locations in
the brain
Anatomy of the Brain

• Cerebral Hemispheres
have four main
– Frontal
– Parietal
– Temporal
– Occipital
The methods of Cognitive Neuroscience

1. Neuroanatomy
2. Neurophysiology
3. Neurology
4. Functional Neurosurgery
5. Cognitive Psychology
6. Computer Modelling
7. Converging Methods
1. Neuroanatomy
• Studies the nervous system’s structure
• Describes how the parts are connected
• Descriptions can be made at many levels
• For the neuroanatomist investigations occur at two levels:
– gross neuroanatomy: general structures and connections
– fine neuroanatomy: main task is to desribe componenents of individual
• Histology is the study of tissue structure through
dissection, and is essential for neuroanatomists to know
1. Neuroanatomy

• Primary concern of neuroanatomy is the pattern of

connectivity in the nervous system that allows information
to get from one site to another
– problem made difficult by:
• fact that neurons are not wired together in a simple fashion
• often innervated with many neurons
• Solution: Refinement of New Stains
• stains for cell bodies
• stains for axons
• have the characteristic that they are absorbed from specific chemicals
and therefore ‘colour’ specific targets
1. Neuroanatomy

• Interested in describing
the structure of different
• Neurons are
heterogeneous, varying
in shape and size
2. Neurophysiology
• Structure is closely tied to function
• We cannot understand brain function from neuroanatomy alone
• Neural function depends on electrochemical processes and numerous
techniques exist to measure and manipulate neuron activity
• Some record cell activity in passive or active conditions and other
manipulate activity by electrical stimulation or chemical induction
a. Electrical Stimulation
b. Single Cell Recording
c. Lesions
A. Electrical Stimulation
• Early insights to cortical organisation were made by directly
stimulating the cortex of awake humans undergoing neurosurgery
• Pioneers, Penfield & jaspers (1954) explored the effect of small
electrical currents applied to the cortical surface
Stimulation of the Stimulation of the
motor cortex: somatosensory area:
movement somatic sensation
B. Single-Cell Recording
• The most important technological
advance in neurophysiology has been
the development of methods to record
directly the activity of single neurons
in laboratory animals.
• An thin electrode is inserted into an
animal’s brain (brain does not hurt!)
• The primary goal of single cell
recording experiments is to determine
experimental manipulations that
produce a consistent change in the
response rate of a single neuron
C. Lesions
• Neurophysiologists have studied how behaviour is
altered by selectively removing one or more of
brain components.
• Logic: if a brain structure contributes to a task
then removing that structure should impair
performance in that task.
• Human cannot be subjected to such procedures, so
human neuropsychology requires patients with
naturally occuring lesions.
MRI scan of a normal and lesioned brain
3. Neurology
• Human pathology has provided key insights to the relation between the brain
and behaviour
• Postmortem studies by early neurologists such as Broca and Wernicke were
instrumental in linking the left hemisphere with language functions
• By selecting patients with a single neurological impairment, we can best link
brain structures to specific cognitive functions.
• Sometimes patients have diffused damage and then conclusions are harder to
– Structural imaging of neurological damage (CT) helps define the damage (advanced
method of x-ray studies)
• Causes of Neurological Disorders
– vascular disorders (ie strokes)
– tumours
– degenerative and infectious diseases (MS, Huntington’s Disease)
– trauma
• Functional Neurosurgery (lobectomy)
Phineas Gage Case
• Most famous patient who survived severe brain damage
• He was a railway construction worker who got injured by
an accidental explosion
• Severe personality change after the accident
4. Cognitive Psychology
• Cognitive Psychology assumes that our perceptions, thoughts and
actions depend on internal transformations or computations
– Mental Representation and Transformations
• information processing depends on internal representation
» ball rolls down a hill -pictorial representation better than one that
encompasses the laws of physics
• mental representations undergo transformations
» imagine two letters presented in a screen one vertical the other one
rotated in order to decide if they are the same or different you
transform them to be into the same position
– Constrains on Information Processing
• exploring the limitation in task performance
» Stroop task
5. Computer Modelling
• Models are explicit
– they can be analysed in detail, the way the computer represents the
process must be completely specified
• Representation in Computer Models
– models differ greatly in their representations (ie. symbolic of object
recognition would have units that represent visual features such as
• Models lead to Testable Predictions
• Limitations with computer models
– radically simplified and limited in their scope
– some of their requirements come in contrast with what we know about
living organisms
– restricted to narrow problems
– modelling often also occurs in isolation to current theories
7. Converging Methods

• Cognitive Deficits Following Brain Damage

• Single and Double Dissociations
• Groups versus individuals
• Imaging the Healthy Brain
Single and Double Dissociations
• Single dissociation
• Two groups differ on one critical behavioral task
• One group has a particular brain lesion, the other doesn't (the
other group is usually a control group who is considered healthy
and without any known brain abnormality)
• We then tentatively conclude that the difference on the
behavioral task is due to the brain lesion
• This, in turn, suggests that the brain region that is lesioned
probably was responsible for some aspect of the behavior being
• However, this connection is not guaranteed to be the case
Single and Double Dissociations
• Problems with interpreting a single dissociation:
• The task measuring the behaviour may not be sensitive to the true
underlying behaviour that is disrupted
• The task may reflect something similar to, or a derivative of, or part of
the real behaviour that brain region is involved in, but it may not be a
completely accurate measure
• The behavioural change, though apparently narrow in scope, may be
part of a broader behavioural change that we haven't yet identified
• The lesioned brain area may also affect other brain areas responsible
for producing this and related behaviours
Single and Double Dissociations
• Double dissociation
• Two groups differ, in different ways, on two different behavioral tasks
• Usually, the two groups each have different types of brain lesions
• For example, one patient with Broca's area damaged and another patient with
Wernicke's area damaged
• The first patient shows difficulty producing speech, while speech
comprehension is apparently normal
• The second patient shows difficulty comprehending speech, while speech
production is apparently normal
• We conclude, fairly confidently, that Broca's area is responsible for speech
production while Wernicke's area is responsible for speech comprehension
Single and Double Dissociations
• Double dissociations are more powerful than single dissociations
because we can isolate fairly specific behaviours that change with one
type of lesion but don't change with a different type of lesion

• The problems with the task (how sensitive it is to the actual

behavioural change) are still a concern, but we are more confident
with conclusions about brain localization when there are double
Groups versus individuals
• Individual case studies
• Study one individual carefully with a known brain deficit
• If there is a specific behavioural deficit (after careful testing), it can be correlated
with the known brain deficit
• And if two case studies are compared, each with different lesions, and double
dissociations are found, we have strong confirmation for the link between behavior
and brain region
• We are, of course, concerned that one individual case study may not reflect a larger
• If you are familiar with statistical analysis, you should know that one research subject
(N=1) is not very useful in statistical analyses
• We cannot know for sure that the behavioural deviations from normality are due to the
brain deficit and not just because this person was different (with or without the brain
Groups versus individuals
• Group studies
• In this approach, we compare groups of people with similar brain deficits and
determine if they show a consistent pattern of behavioural deficits
• This minimizes the chance that individual differences are masking the results of
brain damage
• The bottom row shows the proportion of overlap for a given brain region
• So we would be fairly confident that the areas of highest overlap were most likely
involved in producing the behavioral deficit

• Comparing across brains is not trivial, however, because of individual variation

• To accomplish this, individual brains are matched to a "standard" brain
• The common technique is to use the Talairach brain--the brain of a French woman
• After matching certain landmark features, the image of a brain is distorted until it matches
the Talairach brain

• Then all the brains are compared from this common, standard brain image
Imaging the Healthy Brain
• Transcranial magnetic stimulation (TMS)
• The goal of this technique is to intentionally induce a temporary "lesion"
• As far as we know, the brain is not damaged in any way, but a region is
temporarily deactivated
• A strong electrical signal is sent to a region of the scalp
• We don't exactly know how this works, but it seems to disrupt neural function
• So for a very brief period of time, the behaviours associated with the focus of the
TMS should be impaired

• There is some control over the location of the "lesion," but the precision is
• The device that administers the electrical pulse is fairly large

• It is usually held in place manually, lacking much precision

Imaging the Healthy Brain
Scalp recordings
• Electroencephalogram (EEG)
• Passively measure electrical activity from neurons that reaches the
• Place electrodes on the scalp to record electrical activity
• Hook the electrodes to an amplifier to boost the signal (very little
neurally generated electricity will reach the scalp)
• Have a representation of global neural activity
• Very useful for determining sleep patterns
Imaging the Healthy Brain

• Event-related potential (ERP)

• The development of ERPs is when the EEG became useful as an
experimental tool
• The procedure involves time-locking an EEG recording to the onset
of a particular stimulus or behaviour
• One EEG reading is very noisy; i.e. the electrical signal is very
chaotic and variable
• But if we measure EEG multiple events of the same type, all time-
locked to the onset of the event, and average them together, a smooth
pattern arises
Imaging the Healthy Brain

• Magnetoencephalogram (MEG)
• The methodology of MEG is very similar to the methodology
for ERP
• The sensors for MEG are actually measuring magnetic fields
produced by neurons, not electrical signals
• The inverse problem still exists for MEG, but because there is
less distortion of the magnetic signal than there is for the
electrical signal, the solutions end up being more accurate, on
the whole
• However, this technique is extremely expensive ($1 million for
a reasonably good set-up)
Imaging the Healthy Brain
• Positron-emission tomography (PET)
• Methodology:
• Water labelled with radioactive oxygen, is injected into a subject
• Brain cells require oxygen (and glucose) for energy
• The radioactive oxygen is unstable enough that protons break off and collide
with electrons in the brain
• These collisions are measured by a PET scanner
• With this technique, we do not directly measure neural activity
• It is assumed that the higher concentration of radioactive isotopes reflects
higher neural activity
• The more active a neuron is, the more energy it should need to replenish and
the more likely the radioactive oxygen will enter into that brain region
Imaging the Healthy Brain
• We use the subtraction method to determine relative levels of neural activity
• PET scans are taken separately for two experimental conditions
• The two conditions are identical except for one feature--the behavior being studied
• Then one PET images are subtracted from the other, so the resulting difference
should reflect the defining feature
• So if Task 1 required Processes A, B and C, and Task 2 required Processes A, B, C
and D, the difference between the PET images for Tasks 1 and 2 should reflect the
activity unique to Process D
• One consideration when using PET as an experimental technique is that it
takes 20-45 minutes for the radioactive isotope to get flushed out of the brain
• So each experimental condition takes that long, meaning it is impossible to
compare too many conditions in one PET experiment
Imaging the Healthy Brain
• Functional magnetic resonance imaging (fMRI)
• The BOLD response
• BOLD stands for Blood Oxygen Level Dependent
• What is measured is dependent on the levels of oxygen in the blood for any
local region of the brain
• When oxygen is used by cells, the result is the blood becomes more
• Deoxygenated hemoglobin is more ferromagnetic (the iron in the blood is more
prominent), which is what the MRI scanner can measure
• Basically, fMRI measures the ratio of deoxygenated to oxygenated hemoglobin
Imaging the Healthy Brain
• Subtraction method is one technique also used with fMRI
• Present variations of a task that each differ in one respect
• These differences may be different levels of a single cognitive dimension (e.g.,
different amounts of visual information presented) or they could be completely
different cognitive functions
• Contrast the fMRI signal from these conditions with each other and with the signal
from a control condition, when the extra cognitive function was not present (but
everything else was)
• These subtractive differences are reported as correlating with changes in behavior
• Many different variations are possible with fMRI, unlike PET, because it is not
necessary to wait minutes between conditions
• We can use alternating epochs of a fixed length of time doing each variation of the
Concluding Remarks
• Advances in science are often fueled by
technological developments
• The maturation of cognitive neuroscience as a
scientific field provides a tremendous impetus for
the development of new methods
• The questions we ask are constrained by the
methods available but new research tools are
promoted by the questions we ask.