Lectures 1, 2, 3-Neuroscience cognitive Flashcards
Pictures will be in Gradescope, exam Tuesday.
What is the course overview:(4 categories)
Definition: Cognitive neuroscience explores how the brain enables the human mind, supporting thought and behavior.
Key Topics Covered:
Vision – How the brain processes visual information.
Memory – Neural mechanisms behind storing and recalling information.
Movement – Brain control over voluntary and involuntary actions.
Cognitive Control – Executive functions like decision-making and attention.
Course Focus:
Current Understanding – Overview of major findings in cognitive neuroscience.
Research Methods – Tools used to study the neural basis of cognition (e.g., fMRI, EEG, lesion studies)
What are the goals of this course?
The goals of the course are to prepare you to:
1. Describe the key computational challenges or problems that the brain must solve in the cognitive
domains studied.
2. Identify (on brain diagrams and images) the primary brain areas and networks involved in the
cognitive domains studied (e.g., vision, audition)
3. Describe the major disruption and recording methods used in cognitive neuroscience, explaining
the strengths and weaknesses of each.
4. Explain major cognitive neuroscience findings that have contributed significantly to our
understanding of major cognitive domains.
What is cognitive neuroscience?
- What is cognitive Nueroscience?
- A young science: Historical backdrop
- Neuroanatomy overview.
What is the camera theory? Is it right or wrong?
The brain records the world like a camera, providing accurate(although incomplete) pictures of the world that form the basis of our actions.(WRONG)
What is the brain important for?
To think(Information).
What is the camera theory? Simple definition that he say?
Is creating this sort of hight fidelity representation of the world around us and then using that to guide our actions.
What were the observations from the pasta? His explanation?
there was no noodles in the bowl, Decreasing zoom out empty space.
Almost everyone- will draw it out from the frame.
they sort of extrapolate beyond the edge they saw. Phenom called boundary extension.
What is it called when we saw the spaguety we drew zoom out ?
Boundary extension
What is boundary extension?
Boundary extension reflects cognitive processes that extrapolate beyond the boundaries in the stimulus to provide the basis for planning eye-movements and other actions in the environment (Intraub et al., 1989)
“Demonstration the brain is like a recording, it’s like a camera that takes a snapshot of the world, then use vertical images of the world to guide our behavior.” is this true or not? What does it mean?
But, made completely different image is not.It is in a very different spatial scale. It is also, different for each person in the same way. it is constant thought across multiple people.
What is the hypothesis 2 since the first one is incorrect?
When we view the world, we have a limited picture of the world around us, but there’s enough regulators in the environment, that we can generate prediction of what is beyond or view.–It is useful to guide our movement to make a life plan.
What illustration reveals?(1)
1) Reality is constructed: We do not perceive the world vertically( What we’re getting, you know , our internal representation of the world is not just like a camera. its not a snapshot of the world. its a constructive representation of the world in the mind)
1a) Constructive not just a picture in the mind, can be different like a optical illusion.
1b) It is useful for behavior, but does not necessarily has to be a vertically response to be a reality.
What illustration reveals?(2)(3)
The mind/brain influences perception: Our minds/brains imposes structures(via cognitive process) on the stimulation we receive contributing to the really we experience.( Mind brain intencherchangeble, mind imposing the structure or representation, systematically altered)( inputs and outputs)
Universally: Despite clear individual differences our mind function similirarty( Everyone makes the same mistake).
What is cognitisim(3)
The brain is an information-processing system
* Information processing: Mental representations are transformed by cognitive
functions to support perception, thought, action, and so on
* The brain instantiates multiple parallel and sequential cognitive functions that
mediate between input and output
* To fully understand the brain, we must understand the cognitive functions itinstantiates and how it does so.
What is the picture of the cognition.
Stimulation-> cognitivism-> response
What is the overall idea of the brain?Cognitivism perspective
The idea is that the brain instates cognitive functions that transform, say, in this case, sensory inputs, to generate memory representations and representations that guide behavior.” Cognitivism perspective, logical function.
“Like we are not going to study the brain by looking at neurons, and its patterns of electrical activity”(look at the informed process)
In contrast to cognitivism: Reductionism
“the idea that the most fundamental layer of nature holds the explanation for all the features of the outer, higher layers” (Williams, 1997)”
(goes down)
Cognition
Biology
Chemistry
Physics
MATH
What does Reductism say?
That we can study cognition, but really one we have a deep understanding how neurons word do
reductionism vs cognitivism what is the difference between them?
If brains are, after all, just
assemblies of cells, then once we
truly understand every facet of all
cell function, the principles of brain
function will be evident.
Even if we did know about all the
synapses, all the transmitters, all the
channels, all the response patterns
for each cell and so forth, we would
still not know how an animal sees
and smells and walks”.
While this reductionist perspective suggested, okay, you might be able to find some fundamental level and describe everything on that level, the cognitive perspective is this idea that there’s multiple levels of which we can understand the brain.
What is cognitivism and Neuroscience?
Dfferent branches of neuroscience represent different levels
of analysis of the brain:
* Molecular neuroscience(Neurobiology) seeks to understand the brain at the
level of its biochemical processes
* Cognitive neuroscience seeks to understand the brain at the
level of its information processe.
We need to think about the cognitive perspective. Since it will not tell us nothing, need to look at it as abstract as possible information processing devise. Both are necessary in understanding the human brain.
What is the computer theory for the brain?
We cannot understand a computer without understanding the
software it is running.
We cannot understand the human brain without understanding its
cognitive functions.
Understanding cognition. (abstaract)
What is the mathematical theory associated with the brain?
A cognitive function is something that takes some inputs, performs a transformation on them, and then generate an output
What is the pattern of light example?
But as you’ll see, we can characterize in terms of a representation. So there’s a pattern of neural activity on the retina. But we can think of it as a representation of a pattern of light.
And that pattern of neural activity gets transformed in some way that ultimately leads to, for example, your ability to identify who a person is based on seeing their face. What’s happening is that patterns of neural activity are being transformed along these processing stages until you get to the pattern that’s associated with a person’s facial identity.
And at a more abstract level, we can think of it as patterns or types of representations, so like patterns of light, patterns of edges, patterns of the 3D features of the face, and then ultimately the facial identity.
What is the function of the brain?
What is the function of the brain? The function of the brain is not to conduct electricity across from neuron to neuron or send neurotransmitters from one part of the brain to the other. That’s not the right way to think about the function of the brain. We really want to understand the brain. We need to think about its function in terms of information processing.
What electrical function does the brain do?
It’s moving electricity all around from neuron to neuron. For information processing
Why is cognitive a young science?
More than a century modern research on the brain and cognition, neural theory, and cognitive theory.
Neural theory (neuroanatomy, physiology)
* Cognitive theory (psychology, cognitive science
The integration of neural and cognitive theory is
RECENT
* The first Cognitive Neuropsychology textbook: 1998!
Cognitive neuoscience , young field, cognition long time, basic of neuroscience. Both research integrated together is fairly recent.
Where is the mind History(4)
1)Cardiocentrism: Mind located in the heart poetic
2) Ventricular localization:So it was a major focus on fluids as the key sort of underlying physiological substrate for lots of different things, including the mind. And they thought, oh, here’s this area that has a lot of fluid inside of the head. The head seems to be important for cognition. Maybe this fluid is what underlies the mind.
Very detailed theories. Not all true.
3)cerebral localization:
Tissue in the brain is responsible for thinking since we had the same ventricular as other animals and supposedly we are smarter and different thus it has to be the brain
4) cortical localization:
People had accidents that were very specific to some type or regions.
What is cardiotrism?
Cardiocentrism: the belief that the heart
is the seat of intellectual and perceptual
functions
* Aristotle (384-322 B.C.)
* “broken heart”, “heartfelt thanks”, “learn by
heart”
What is Ventricular localization? And the 3 parts of it
1)Ventricular localization: the fluid-filled
ventricles are the location of cognition:
2)Perception/imagination: lateral ventricle
3)Cognition: third ventricle
4)Memory: fourth ventricle
Widespread view from the 1st century through
the Middle Ages and Renaissance
What is cerebral localization?
Realization that the ventricles are unlikely to be the seat of higher mental functions because the ventricles of humans are not
dissimilar from those of animals with clearly inferior mental abilities
* The localization of respiration to the medulla and sensory and motor functions to the dorsal and ventral roots of the spinal cord led
to the idea that other parts of the brain might support other, higher-level mental functions
Who was Gall(German physician
and anatomist): 1758-1828? And what did he Believe?
Cortical Localization:
* The brain is composed of as many organs as there
are mental faculties (aka: cognitive functions)
Phrenology:
* These mental organs vary in size and their size
affects the shape of the skull
* One can study a person’s character and faculties by
studying the external configuration of the skull
* Sought evidence by studying the skulls of people
with exceptional traits: poets, statesmen, people
with mental illness, etc.
What is phrenology?
Phrenology:
* Correct about cortical localization of function
* But wrong about evidence from skull shapes
It turned out to be a pseudoscience, but at the time there were
a lot of adherence to this idea that
that there are different specialized functions in different parts of the brain and that
That just like in the way that if you use a muscle a lot, the muscle will grow. That if you use a particular cognitive function a lot, that part of the brain will grow.
And actually that you could then even tell this by looking at the surface of the skull. That there would be bumps in the skull, that would tell you something about people’s mental faculties because they have an enlarged part of the brain that’s involved in memory or something like that.
turned out not to be true, but
This person, Franz Gahl, developed a whole theory about this. It became very popular. There were journals and societies dedicated to it.
Now, it turned out that he was wrong about this, but there was one sort of element of this idea of phrenology that actually turned out to be right, which is that there is
functional specialization of the brain. Different regions of the brain do have different specialized functions, right?
It’s just that those regions don’t get larger if you are better at that function.
You definitely can’t see anything on the skull, right? You can’t tell if somebody has a good memory by likegetting some information about the shape of their skull.
Competing views on the neural instantiation
of cognition? What are the 2 different concepts?
Functional localization:
Specific cortical areas perform specific cognitive functions
* There may be interactions between cortical areas, but these interactions
are sufficiently limited, such that we can usefully understand the brain by
identifying its cognitive functions and their localization in the brain
And the idea is that of course there are interactions across brain regions and we need to, you can’t just like, we can’t fully understand the brain by just studying.
each brain region of isolation. At some point we need to understand how they interact with each other. But there’s enough localization of function. Study each section individuality.
Holism:
The cortex is a dynamic whole which is more than the sum of its
parts
* Music cannot be understood as individual notes
* Squareness cannot be understood by individual lines
* The brain cannot be understood via decomposition into local areas and
component functions
The contrast to this is something like holism where
The idea here is that the neural basis of cognition is so widespread and so dynamic in the brain.
That
You can’t possibly go in and study one brain region and try to characterize what its cognitive function is. It’s so interdependent on every other region of the brain that you’re just misleading yourself. That’s like trying to understand music by studying the individual notes or whatever, not understanding how the notes work together to form a piece of music that gives an aesthetic experience.
You’re missing the sort of bigger picture by focusing on individual regions.
Who was Paul Broca:1824-1880
French physician
* Tested the hypothesis that
language is a faculty of the
anterior frontal lobes
Tan (Leborgne): extreme difficulty in
speaking but no paralysis of lips or
tongue, and no language comprehension
difficulties
* Autopsy revealed damage to the “3rd
convolution of the frontal lobe of the left
hemisphere” (which we now call the
inferior frontal gyrus)
* Turned the tide in favor of functional
localization (as opposed to holism)
And he had no difficulty with language comprehension. It was clear that you could talk to him, he could follow directions, but he couldn’t speak. He had this difficulty generating speech.
Related debate: Neurons as continuous cytoplasm or independent units?
Fused neural networks (Golgi)
Independent units (Ramon y Cajal)?
If all neurons are fused, how can there be localization of function, as suggested by Broca’s findings?You know, the microscopes weren’t as powerful, the techniques for isolating cells weren’t as good. So it was actually really difficult to try to resolve this debate. It turned out that the key piece of evidence for resolving this debate came from Golgi.
Turning point: Silver nitrate
So Golgi had been arguing in favor of this like Fuse network idea and he was experimenting with different types of stains so he had these like you know dishes of slices of neural tissue that he was staining and trying to characterize the structure of the neurons and he just hit upon this particular stain that would get picked up by specific neurons. It’s actually not fully understood why. Specific neurons would pick up the stain and it would diffuse through that neuron
And it would give you like a very clear picture of that neuron and it wouldn’t go to any other neuron.
And so a lot of people saw these results and thought. Well, this is very clear evidence that all the neurons are not fused together because there’s clearly like an endpoint to where this goes. It doesn’t just fuse to all the other cells. It seems to be restricted to this one cell. It suggests that it’s an independent unit.
Interestingly, Golgi himself was not convinced of this idea from this evidence. He went on to win a Nobel Prize, and even in his Nobel Prize acceptance speech, he still argued for the fused neural network idea. But most of the rest of the neurophysiology community was convinced and then of course we then later learned that indeed that was correct and it sort of was the starting point of what’s called the neuron doctrine. The idea that the sort of most basic element of the brain is the neuron and each neuron is an independent
Silver nitrate stain: Stains random neurons.
Discovered by Golgi (1873) in a kitchen,
working by candlelight
* Convinced most (but not Golgi himself)
that neurons are not continuous and
connected—instead, they are independent
units
* The Neuron Doctrine
* Neurons are separate physiological units
* Electricity travels in one direction down a
neuron
Purkinje cell (cerebellum)
How Do neurotransmitter communicate(Draw picture, then explain)
Presynaptic neuron receives signals->
Moves current down an axon->
Triggers release of neurotransmitter at the synapse->
Passes signal to post-synaptic neuron
Addition and Subtraction:
Excitatory signals (EPSP): increase the probability that the
post-synaptic neuron will fire (generate a signal)
Inhibitory signals: (IPSP) decrease the probability that the
post-synaptic neuron will fire (generate a signal)
Electrical signals are the basis of information processing in
the brain
what’s happening when neurons are
communicating with one another is that
Neurons send signals to one another
either synapses by releasing neurotransmitters.
One neuron releases a neurotransmitter and the neurotransmitter lands on a downstream neuron.
It changes the likelihood that that downstream neuron will itself send signals to other neurons. So at a very simple level, we can think of this as either
exciting the downstream neuron or inhibiting the downstream neuron. So it excites the downstream neuron. It’s increasing the probability that the downstream neuron
will fire an active potential and then some electrical activity will travel down the downstream neuron and it will release neurotransmitters that it’s synapses.
If it’s inhibitory, then it decreases the probability that the downstream neuron
will fire an actual potential.
So at the simplest level, we can think about it as like neurons are sending signals to one another that basically increase or decrease the probability that the next neuron will then send some other signals to another set of neurons.
So it’s like, well, pluses and minuses. You have a bunch of neurons doing this, and it determines when neurons are finding their action potentials.
So this is the basis of information processing in the brain.
Neural computation
100 billion neurons in
the human brain
* typical neuron has
1,000-10,000 synapses
* ~60-100 trillion
synapses
* A LOT of computational
capacity!
We talk about neural representations when we talk about sending signals, when we go on information processing. It’s all really instantiated in neurons sending signals to one another. And as we’re going to see, we can get a picture of that by recording the actual potentials of neurons. It tells us when they’re sending signals to other neurons.
or using other methods to indirectly tell us about when neurons are firing action potentials.
Okay, so
What is neurotomy?
earning neuroanatomy is analogous to learning a language or
geography; it is a slow process of building inch by inch and learning by
repetition.” (DeArmon, Fusco & Dewey. Structure of the Human Brain.
1989)
Directional terms
Directional terms
* Dorsal or superior = top
* Ventral or inferior = bottom
* Anterior or rostral = front
* Posterior or caudal = back
* Medial = middle
* Lateral = sides
Note that these are relative terms
* Bilateral = on both sides (hemispheres)
* Ipsilateral = on the same side (hemisphere)
* Contralateral = on the other sid
What are the names of the brain surfaces? Watch picture
Dorsal or superior means at the top.
And by the way, these directional terms are relative terms, right? So we can say…
Like this, you know the superior part of the brain overall So it’s like the very top of the brain or I could say like the superior part of the temporal lobes We’ll see what that is soon, which would be like actually like halfway up the brain, right? This is like the top part of the temporal lobes
Ventral or inferior means the bottom.
Anterior or rostral means the front and a posterior or caudal means the back.
Medial means in the middle and lateral means on the sides. Again, these are relative.
I could talk about like the medial aspect of a particular gyrus or something like that, right?
doesn’t necessarily mean the middle of the brain, it means like the middle, you know, the part of that gyrus that’s closer to the middle. here are other terms that describe the laterality.
So you could say bilateral, meaning on both sides of the brain. You could say ipsilateral, meaning on the same side. So this will come up when we talk about like lesions. So you could think about like the effect of a lesion on
cognition or on like say motor control for example and you can talk about that as being ipsilateral or contralateral. So for example if you have a lesion to right motor cortex it will result in difficulty controlling the left side of your body right so that that’s a contralateral lesion right or the impairment
is contralateral to the lesion, right? You can say it either way, right? Whereas if’s lateral would mean on the same side.
How many vasculature are there?
400 miles of blood vessels
what are the ventricles( Do pictures)
Ventricles contain cerebrospinal fluid
(CSF), produced by blood vessels in the
ventricular wall
* CSF is also present between the brain
and skull to protect the brain, like a
shock absorber
What are the key structures
Thalamus
* Limbic system
* Cerebral cortex
What is the thalamus?(look picture)
Central relay for all sensory
and motor information to
and from cortex
* Lateral geniculate nucleus:
visual information
* Medial geniculate nucleus:
auditory information
* Ventral posterior nucleus:
somatosensory information
And then also there’s another part of the thalamus that is a relay station for auditory processing, the medial geniculate nucleus. So we won’t talk much about what happens in the thalamus. For our purposes, it’s really basically just a path, it’s a relay station along the pathway.
What is the limbic system?picture
The limbic system is this set of structures that’s a realm, the thalamus.
the hippocampus, the amygdala, and the cingulate jar. So all of these are going to come up in this course. The hippocampus is particularly important. We’re going to learn
about this when we learn about memory. It’s really a key structure for memory.
Hippocampus
* Medial
temporal lobe
* Key for
memory
formation
* Amygdala
* emotion
* Cingulate gyrus
* Motivation,
error detection
You can see on a coronal slice of the brain, you can see the hippocampus in this medial part of the temporal lobes. It’s an eventual part of the brain and it’s medial and it has a sort of like coiled up structure.
The amagal is this sort of this like ball shape structure that’s right in front of the hippocampus. It’s really important for emotion.
And then the cingulate gyrus is this large gyrus that goes around this white matter structure which you’ll see here called the corpus venosum that connects the two hemispheres.
And we’re going to see that the singular chirus is important for controlling our behavior. We’re going to see how it’s involved in monitoring when we’re making mistakes.
and so on.
What is the Cerebral cortex? look at picture
And then the white matter is all of the connections between the neurons.
It’s all the axons. It’s all the wiring of the brain.
And it’s covered in lipids. It’s fatty. That’s why it shows up with white in these images.
and
It’s the part of the brain where neurons are sending signals to one another, axons to one another, sort of a wiring that connects different neurons to one another
What are lines between brain?look pic
for (singular-gyrus): protruding
* Sulci (singular=sulcus): furrows
* Some large sulci are called fissures
White-matter tracts (axon bundles)
Commissures are white matter tracts that
connect the two hemispheres:
* Anterior commissure
* Corpus callosum (callosal commissure)
* Posterior commissure
* Within-hemisphere tracts
connect gray matter regions
within a hemisphere
what is Disruption/stimulation approach?
So the first is about disruption and stimulation methods where where, either there’s, like, a naturally occurring lesion, like a stroke or something, just disrupted, some, some functioning in the brain, and then you’re trying to characterize how did that affect cognition. Or there’s actually as we’re gonna see, there’s methods for directly going in and sort of stimulating the brain and seeing what happens when specific brain regions are stimulated.
Disruption/stimulation approach:
* Disruption/stimulation of a specific brain region can disrupt/enhance specific cognitive
functions
* We can use experimental tasks to determine which cognitive functions are impaired (or enhanced) when a specific brain region is disrupted (or stimulated)
* This approach can determine which cognitive functions are supported by the disrupted/stimulated brain region
So this is a really, powerful method for trying to understand which regions of the brain have a causal function in causal relationship to a specific cognitive function.
Mapping the brain cognition* Neural-activity recording approach:
- Design experimental tasks to isolate a
cognitive function - Measure brain activity and determine which
brain regions are activated by the task
See the pictures of Disruption/stimulation approach:, and Neural-activity recording approach:
Disruption: Brain lesions( be able to identify pictures)
Stroke
* Stroke in the middle cerebral artery,
damaging most of the frontal lobe) and
the superior temporal gyrus
* Tumor
* Tumor in ventromedial right hemisphere
* Degenerative diseases (Alzheimer’s,
Parkinson’s, Huntington’s, Pick’s, and
others)
* Frontal and temporal lobe atrophy
(gradual loss of grey and white matter,
more distributed than damage from
stroke)
* Trauma
* Ventral frontal head injury
Provided an example on how to mainipulate cognition?
So I’m gonna show people images in the scanner. Some of them will be faces, some of them will be other things. And I wanna see what effect that has on the brain activity, what region of the brain is activated when people are perceiving faces. So notice the direction of the sort of causal arrow here. Right?
So in the first thing that we talked about, disruption, you start with some manipulation of the brain. Right? So naturally occurring, lesion of the brain or a direct experimental manipulation of the brain. And you assess the effect on cognition. Right?
So that that’s a situation in which you can see the causal effect of this brain region on this cognitive function. Right? In the recording approach, you’re manipulating cognition and you’re looking at the effect on brain activities. Right? So your starting point is to manipulate some task and see what happens in the brain.
Right? Now there’s an important thing to keep in mind, which is when you’re doing this recording approach, you you can’t say for certain with certainty that this brain region say a certain brain region is activated during face perception, for example. You don’t know for sure that this brain region has a has a causal role in face perception. Right? Because you don’t know, like, maybe it’s the case that you thought it was about face perception, but it was really something else.
What are the tree types of inference in brain-functioning mapping?
1) Association
2) Dissociation
3) double dissociation
Why association can be wrong?
Now there’s an important thing to keep in mind, which is when you’re doing this recording approach, you you can’t say for certain with certainty that this brain region say a certain brain region is activated during face perception, for example. You don’t know for sure that this brain region has a has a causal role in face perception. Right? Because you don’t know, like, maybe it’s the case that you thought it was about face perception, but it was really something else.
Like all those in all those photos had some other thing in them that was correlated with faces that you didn’t take into account. Right? And that was causing activation of this brain fusion. Like, they all tend to be brighter photos than the other photos that you showed to the subjects. Right?
Association what do we have?
Damage to brain region 1 is associated
with poor performance on cognitive
function A
* E.g., damage to the occipital lobe is
associated with poor vision
* Inference: brain region 1 is involved
in function A
You need to show that when that brain region, its activity has been disrupted, that the cognitive function is also disrupted
So the first method is showing you, like, the causal role that a brain region has in specific cognitive functions, and the second region is just determining associations between brain regions and cognitive functionsyou can study, say, one person or multiple individuals that have a damage to a particular brain region, say, brain region 1, and, you find that when you administer some tasks, it’s like a visual perception task, they’re really bad on that task. Their accuracy is, like, 20%, and healthy individuals have an accuracy of, like, 100%.
Right? So what you found is that you found an association. Right? So you found that damage to brain region 1 is associated with worse performance on the task that you administered. Right?
For example, maybe damage to the occipital lobe is associated with poor, performance on vision test. Right? Okay. So this is one type of finding. It’s it’s showing you that this brain region is involved in this task.
Right? But what’s lacking is a degree of specificity. Right? And the way to think about this is maybe it’s the case that damage to this brain region makes you bad on lots of different types of tests. Right?
Like, for example, let’s say it was, like, the occipital lobe is, like, is is important for vision. Okay? And so let’s say what you did was you administered, like, a math exam. And then the the subjects were really bad at doing the math exam. And you said, how much is the occipital lobe?
Makes people worse in math. Right? But then really, what’s going on, and if you did more studies, you realize, well, actually, they just have bad vision, and they can’t do, like, any exam. Right? So, yeah, it’s like, has a causal role in doing, like, math exams where you have to, like, read something and and do an exam, but it’s not really about math.
Right? That’s not the specific association that you would you would wanna draw. It’s really vision. And so one way in which you can rule out these kinds of explanations, these more general explanations, is to find dissociations.
Lesion-deficit correlation?
- Identify/generate brain lesion
- Evaluate cognitive functions
- Across multiple individuals (or studies) determine if
there is a systematic relationship between a cognitive
function and a brain area
Drawing inferences from
disruption/stimulation methods
Double dissociation
* Requires two groups or subjects with damage to
different brain regions
* Group 1: damage to region 1 is associated with poor
performance on function A but not function B
* Group 2: damage to region 2 is associated with poor
performance on function B but not function A
* E.g., damage to the fusiform gyrus is associated with
poor perception of faces but not places, and damage to
the parahippocampal gyrus is associated with poor
perception of places but not faces
* Inference: brain region 1 is involved in function A
but not function B, and brain region 2 is involved
in function B but not function A
* Strong evidence for more specific brain-function
mappings and very strong evidence against more
general explanations (e.g., general deficits in
intelligence, attention, etc.)
Right? That’s not the specific association that you would you would wanna draw. It’s really vision. And so one way in which you can rule out these kinds of explanations, these more general explanations, is to find dissociations. So you can show that, like, okay, well, the, the subjects with, damage to brain region 1 have poor performance on task a, but they’re fine on task b.
ike, there’s a region of the visual system that’s really important for face perception. And when people have damage to that brain region, they have a specific impairment in face perception. Right?
So you would find the damage to the fusiform gyrus is associated with poor perception of faces, but not objects or other types of visual stimuli. Right? And we’re gonna we’re gonna have, like, a whole lecture where we get we talk about, like, how people figured out that this brain region is specifically involved in face perception. But just at a broad level, just to give you an idea of what this means. Right?
So you’re like, you know, you’re trying to find other tasks, so it’s like, oh, I think it’s about face perception. You’re trying to find other tasks that are similar to face perception as possible. Right? They’re visual stimuli. They’re doing the same kind of responses, but yet if you can find that they are still doing this task just fine, right, but yet they’re bad on the taste task, that gives you more confidence that you sort of identified the specific cognitive function that’s affected here.
Okay. So, so an association’s result is one where you found an association between a performance on a particular task that you think is tapping into some cognitive functio
ecause you’ve only tested this one task, so you haven’t ruled out other sort of more general explanations. Like, for example, maybe the reports on every single task you give them. Right? There’s, like, ban on everything you give them. Right?
What did we used to do?
So, classically, the the most common way in which people studied the effect of disruption of brain activity on cognition was by studying naturally occurring lesions of the brain. This still this is still a, you know, a very common method in the field, but now we have a lot more sort of technologies that allow us to, to go beyond just the study of naturally occurring lesions
what is double association?
Double dissociation
* Requires two groups or subjects with damage to
different brain regions
* Group 1: damage to region 1 is associated with poor
performance on function A but not function B
* Group 2: damage to region 2 is associated with poor
performance on function B but not function A
* E.g., damage to the fusiform gyrus is associated with
poor perception of faces but not places, and damage to
the parahippocampal gyrus is associated with poor
perception of places but not faces
* Inference: brain region 1 is involved in function A
but not function B, and brain region 2 is involved
in function B but not function A
* Strong evidence for more specific brain-function
mappings and very strong evidence against more
general explanations (e.g., general deficits in
intelligence, attention, etc.)
nd the idea here is that, you can you’re looking for a group of patients that have damage to one particular brain region. And they’re, they’re bad at task a, they’re good at task b. Alright. So let’s say they’re bad at face perception, but they’re good at object perception. And then you wanna find another group of patients that that have damage to a different brain region where test a their test a performance is fine, or their face perception is fine, but they’re worse on the other thing, right, test b.
Right? So they’re worse on, like, object perception in this example. Right? So let’s go to double association. You found 2 different patterns of of dissociation that go in the opposite direction, right, in 2 different groups of patients that have different brain functions.
Right? This is the strongest type of of finding that you can get, because it really shows you that you found something that’s highly specific, highly specific association between these brain regions and these cognitive functions. Right? Just to give you an example of, like, why this might be important, let’s say you found this first result. You found this dissociation.
You found that, okay. Patients who have damage to the fusiform gyrus are worse on face perception, but they’re fine on object perception. So I found this dissociation. So I think there’s something specific about face perception in the ang in the fusiform gyrus. Right?
But what if what’s going on here is that, like, let’s say you tested other patients that have damage to other brain regions, and you find that they also look like this. Right? And then what you realize is that you created a test that you didn’t you didn’t you weren’t aware of this, but the face task is just more difficult than the object task. So, like, damage to any brain region is just gonna make people worse at any challenging task. Right?
Like, anytime you give them something that’s a little bit challenging, they’re gonna do worse because they just have difficulty concentrating or something like that. Right? So you thought it was something specific, but it was something more general, just like a task difficulty thing. Right? So you can rule that explanation out with a double dissociation
Drawing inferences from
disruption/stimulation methods
Similar logic applies to brain
recording methods
* E.g., perception of faces but not
places is associated with strong
activation of the fusiform gyrus,
and perception of places but not
faces is associated with strong
activation of the parahippocampal
gyrus
d
Lesion-deficit correlation: Characterizing the
lesion
Many ways of characterizing lesions
* Gray matter
* Presence/absence of lesion
* Thickness
* White matter
* Presence/absence of lesion
* Thickness
And then the white matter is all these sort of fibral fiber bundles that are connecting all the different brain regions to one one another. It’s like the wiring of the brain.
We’re gonna see them that can characterize the degree of the lesions to break to the gray matter or the degree of the lesions to the white matter. And both of these techniques that we’re gonna focus on use, MRI magnetic resonance imaging.
Structural MRI(look at powerpoint)
We’re gonna see them that can characterize the degree of the lesions to break to the gray matter or the degree of the lesions to the white matter. And both of these techniques that we’re gonna focus on use, MRI magnetic resonance imaging. here is 3d.
Measures the difference in water (hydrogen) density in brain
tissue to visualize brain structures (static MRI)
* Spatial resolution is determined by the size of the voxels (3D
volumetric pixels) in the MRI image
* If voxels are 3x3x3 mm, there are ~80,000 voxels in a 3D image
of a human brain
MRI: spatial resolution what is it?
MRI: spatial resolution
* Spatial resolution:
ability to resolve
small differences in
an image
* Spatial resolution in
MRI is determined by
the size of the voxels.
* Only very advanced
machines can obtain
voxels smaller than 1
mm.
* A: 8 mm, B: 4 mm, C:
2mm, D: 1.5 mm, E: 1
mm
This is normiazation.
Brain locations: X,Y,Z coordinates in 3D space,
each unit =1 mm
* X (left-right/lateral-medial) distance from
vertical plane between the hemispheres (mid-
sagittal)
* 0 = medial plane
* Positive = right hemisphere
* Negative = left hemisphere
* Y (posterior-anterior) distance from vertical
plane through the anterior commissures (AC)
* 0 = AC plane
* Positive = anterior
* Negative = posterior
* Z (ventral-dorsal) distance from horizontal
plane through the anterior and posterior
commissures
* 0 = AC-PC plane
* Positive = superior
* Negative = inferior
35, -45, -12?
It was created in Montreal, the Montreal Neurological Institute Coordinate System.
t was created in Montreal, the Montreal Neurological Institute Coordinate System.
Lesion-deficit correlation
By overlaying images of brain lesions for a group of subjects
who all have a similar pattern of cognitive impairment, we can
identify common areas of lesioning across the group
* Bottom row: the color coding in each voxel indicates the
percentage of subjects in the group who have a lesion in
that voxel
WHAT ISMRI: Diffusion tensor imaging?
Parietal lobe lesion and
surrounding axonal fibers
imaged with DTI
* Note the right/left
asymmetry in the fiber
tracts. And so you lose basically, what happens in in your DTI data, diffusion tensor imaging data, those track like, you no longer can trace out those tracks. Basically, it looks like those tracks are missing on the brain. Right? So this is an example of, like, a patient who had, a lesion affecting some of the white matter tracks tracks in one hemisphere, but not the other hemisphere. The point here is that on, on one hemisphere, you see all those bright yellow tracks, and then they’re just missing in the other hemisphere.
That’s where the lesion is.
Strengths and weaknesses of lesion-deficit
correlation
Advantages:
* Powerful method for testing the causal role of brain regions in cognitive
functions
* Reveals brain areas that are NECESSARY for the function
* Double dissociations provide strong evidence for separable
processes (e.g., face perception vs. place perception.
DISAVANATAGES
In humans, these are not “controlled” experiments
* E.g., Damage may not occur in brain areas one would like to investigate. Subjects may have
various cognitive, educational or other differences that complicate interpretation
* Damage is not always narrowly localized, and multiple functions can be
impaired.
* Damage is never identical from person to person
You’re not like the the lesion of the brain was not an experimental manipulation. Right? You didn’t have control over that. Right? And so as a result, you’re Right?
You didn’t have control over that. Right? And so as a result, you know, the lesions can be I mean, the lesions are messy. Right? They’re, like, not in one specific brain region.
WHAT IS CORTICAL STIMULATION
Transcranial magnetic stimulation (TMS)
2. Brain electrodes
1. intra-operative
2. long-term implants
3. temporary implants
What was the the story he said?
Most common in front of the head
Dammage to the occipital lobe.
Cortical stimulation
Cortical stimulation
* For research purposes: cortical stimulation can be used for studying brain-
cognition relationships
* What’s the logic?
* Stimulation affects neural activity of underlying tissue
* If this tissue supports a cognitive function, then the ability to carry out the function should
be affected.
* Researchers can draw inferences about the underlying cognitive functions from
performance on one or more tasks under stimulation conditions
* There are also clinical uses of brain stimulation (we will see some examples)
Transcranial Magnetic Stimulation (TMS
Wire coil encased in insulation
* Electrical current sent through the coil à generates a
magnetic field
* Magnetic field passes through skin and scalp à produces
a physiological current à can increase or decrease
neuronal firing
* The current can impact (disrupt) cognitive processing
* The effects are limited to a local region of ~1-1.5 cubic
cm, allowing researchers to test hypotheses about the
localization of cognitive functions
* It is akin to a temporary, reversible “lesion”
* Advantage: Can be used in healthy volunteers
have agents in skull
Intra-operative stimulation
How can surgeons avoid damaging critical cognitive
areas during brain surgery?
* Map brain regions during awake brain surgery
(e.g., for brain tumor, epilepsy, etc.)
* Apply focal stimulation and ask individual to perform
tasks (e.g., name a picture, count, move hand, etc.)
* Brain lacks pain receptors
* Penfield was an early pioneer using this technique
Left: Intraoperative view of parietal tumor before resection
Above:
* The letter tags mark the tumor boundaries.
* The number tags show areas identified using electrical
stimulations:
* primary somatosensory areas of the hand and fingers (10, 11,
12); language sites (20: counting, 21: naming); calculation
areas (30: multiplication, 31: multiplication and subtraction,
33: subtraction).
An: anterior; P: posterior; M: midline
Implanted electrodes (long term): Deep Brain
Stimulation (DBS
Parkinson’s Disease:
Loss of dopamine producing cells in
substantia nigra (basal ganglia)
Placement of DBS
electrodes in
subthalamic nuclei of
the basal ganglia
Temporarily implanted electrodes
- Subdural arrays of electrode pairs
embedded in plastic strips are placed on
the surface of cortex - Recording (to identify epileptic focus)
(ECOG) - Stimulation (injection of current into
electrodes)
Stimulation: - For clinical purposes: cognitive
mapping used by surgeons to limit
cognitive impact of surgery. - For research purposes: Disruption of
function to understand brain-cognition
relationships
Note that it can be challenging to isolate cognitive functions:
We cannot directly measure a “cognitive function.” Instead, we design
tasks that rely on that cognitive function (e.g., picture naming).
* However, a single task typically involves multiple cognitive functions
* And a single cognitive function can be used in many different tasks
Challenges for understanding brain-cognition relations
Isolating a brain region
– creating disruption or measuring activity in a limited brain
region
* Isolating specific cognitive functions
– Comparing tasks that are matched for all functions but the
one under investigation (experimental and control tasks)
Disruption/stimulation techniques: Strengths
and weaknesses
Disruption/stimulation techniques: Strengths
and weaknesses
* Extremely powerful for
hypothesis testing, esp.
to test if an area is
necessary for a
cognitive function
* Limitations in terms of
available neural
locations, invasiveness,
spatial resolution,
temporal resolution,
etc.
What do we look for every time?
Brain activity
Brain activity” is neurons firing action potentials
* Action potentials cause the release of
neurotransmitters at synapses
* This is the primary way in which neurons send signals
to one another
* Communication between neurons is the basis of
information processing in the brain
* All measures of brain activity are either directly or
indirectly related to neural firing
* Directly: e.g., electrical recording from single neurons
* Indirectly: e.g., changes in blood oxygenation (fMRI)
Why measure brain activity?
- Identify what cognitive functions a brain region is associated with
* E.g., face perception
Kanwisher et al. * Fusiform Face Area
F = face stimuli
O = object stimuli
fMRI data
Kanwisher et al., 1997
Why measuring brain activities
Measuring curves. Two thirds (67%) of tuning curves with
a maximum at one extreme showed the minimum at the opposit
Why measure brain activity?
Characterize the functional organization of the brain
* E.g., brain maps
Why measure brain activity?
Identify what cognitive functions a brain region is associated with
* E.g., face perception
2. Identify what information a brain region sends to other parts of the brain.
* E.g., information about specific facial features
3. Characterize the functional organization of the brain
* E.g., brain maps
Recording from neurons
Record the electrical activity of neurons by
invasively placing electrodes inside the skull
* Key measure is firing rate
* Number of actions potentials (spikes) per
second
* Determine what causes the neurons to fire:
* What stimulus features?
* What cognitive tasks?
* If a neuron responds strongly to a specific
stimulus, we infer that the neuron is
involved in processing that stimulus
Recording from neurons( Advantages)
High temporal resolution (milliseconds)
* High spatial resolution (as small as single neurons)
* Can allow for the recording of individual neurons
* Single cells can be highly selective in what they respond to
Recording from neurons(disadvantages)
Invasive
* Samples only a small percentage of neurons in the brain
* Which brain region should I target?
* Cannot demonstrate causal function of neurons in
cognition
* Need disruption methods to address this issue
* What happens when these neurons are disrupted (e.g., lesioned,
turned on/off)?
What is fMRI?
functional Magnetic Resonance Imaging
* Non-invasive technique for indirectly
measuring brain activity
* Uses a strong magnet (often 3 tesla or more)
to detect small changes in the magnetic
properties of blood that occur when the
neurons in a brain region are active
- often 3 teslas or more
What is the process of fmri?
Blood Oxygenation-Level Dependent (BOLD)
signal
* When neurons fire, they require more glucose
and oxygen, which is carried in red blood cells
* When neurons fire, excessive amounts of
oxygen are sent to nearby blood vessels
* More neural firing = more oxygenated
hemoglobin in local blood vessels
* fMRI is sensitive to the ratio of oxygenated to
deoxygenated hemoglobin
* More oxygenated hemoglobin = greater fMRI
signal
What is firm mechanism?
Increased
neural activity
* Increased local blood flow
* Increase in oxygenated
hemoglobin relative to
deoxygenated hemoglobin
* Increased BOLD fMRI
signal
Emodynamic response function (HRF)
The hemodynamic response detected
by fMRI is slow relative to the speed of
neural activity
* Peaks at 4-6 seconds after activity onset
* This lagged response limits the kinds of
experimental questions that can be
addressed with fMRI
* Cannot examine rapid fluctuations in
neural firing
What is FMRI?
When experimental events
are close together in time,
their responses overlap
* Challenging to determine the
amount of activation that can
be attributed to each even
What is FMRI?2
fMRI data are 3D images
(volumes) collected every few
seconds (repetition time)
* Typically collected every 1 to 3
seconds
* Each image is composed of
voxels
* Voxels = volume pixels
* Typically cubes with a length of
1 to 3 mm on each side
How can we use fMRI?
One of the primary ways that fMRI is used
is to identify the cognitive functions that a
brain region is associated with
* E.g., face perception
* In the scanner, subjects see
images of faces, objects, scenes,
etc.
* Compare levels of fMRI signal
across stimulus categories
* Are there selective responses to
certain categories and not
others?
How can we use fMRI?
Our data allow us to reject alternative accounts
of the function of the fusiform face area (area
“FF”) that appeal to visual attention,
subordinate-level classification, or general
processing of any animate or human forms,
demonstrating that this region is selectively
involved in the perception of faces.
How can we use fMRI?
MRI results are often displayed as maps of activation across the brain
* These activation maps show which brain regions respond most strongly to an
experimental condition (e.g., faces)
* fMRI results are simply numbers that indicate the activity levels of voxels.
* Activation maps convert these numbers into color scales.
A R T I C L E S
Cortical surface rendering Slices through the 3D data
fMRI what are the advances?
Non-invasive
* Can examine activity across the whole brain
* Moderately good spatial resolution (millimeters)
fMRI what are the disadvantages?
Low temporal resolution (seconds)
* Cannot examine activity of individual neurons.
* A single voxel may contain 10^5 to 10^6 neurons
* Cannot demonstrate causal function of brain regions in
cognition
* Need disruption methods to address this issue
* What happens when these regions are disrupted (e.g., lesioned,
turned on/off)?
What do EEG and MEG measure?
EG = Electroencephalography
* MEG = Magnetoencephalography
* MEG and EEG are different views of the same neural source
* Electromagnetic signals generated when large groups of neighboring neurons
(10^3 to 10^5) exhibit synchronous activity
What do EEG and MEG measure?
When neurotransmitters are
released at synapses, they allow
charged ions to flow into the
postsynaptic neuron
* This creates a difference in
electric potential along the axis
of the neuron, which can be
detected as a change in voltage
* The electric signal from a single
neuron is very weak and hard to
detect from outside of the head
(non-invasive)
What does EEG and Meg measure?
When the activity of thousands
of neurons fluctuates
synchronously, their electrical
signals sum and can be detected
with electrodes on the scalp
* This is the signal that is detected
by EEG
What do EEG and MEG measure?
Electric currents produce magnetic fields,
which curve around the current
* When the activity of thousands of
neurons fluctuates synchronously, their
associated magnetic fields can be
detected outside of the head
* This is the signal that is detected by
MEG
What do EEG and MEG measure?
These magnetic fields are very weak
(10^-12 tesla) relative to the magnetic
fields generated by the earth, electrical
equipment, and many other
environmental sources
* “a challenging problem akin to listening
for the footsteps of an ant in the middle
of a rock concert”
Magnetically shielded room
Big box. Very expensive
MEG
What is EEG?
EEG can be used to
assess the overall state
of the brain (arousal,
sleep, etc.)
* These measurements
reflect large-scale
changes in overall brain
activity
How can we use EEG and MEG
EEG and MEG can also be used to assess the timing of brain responses to experimental
events
* EEG or MEG responses to an event are averaged over many repeated trials
* These are called event-related responses
How can we use EEG and MEG?
Average event-related MEG signals from
the back of the head contain face-selective
responses at 100 and 170 ms
The picture lecture 3?
- The response at 100 ms is associated with face
categorization - Is this a face?
- The response at 170 ms is associated with face
identification (and also categorization) - Who is this person?
Look at the lecture 3 images that did not load
2
What is egg and meg?
We found a face-selective MEG response
occurring only 100 ms after stimulus onset
(the ‘M100’), 70 ms earlier than previously
reported. Further, the amplitude of this M100
response was correlated with successful
categorization of stimuli as faces, but not with
successful recognition of individual faces,
whereas the previously-described face-
selective ‘M170’ response was correlated with
both processes. These data suggest that face
processing proceeds through two stages:
an initial stage of face categorization, and a
later stage at which the identity of the
individual face is extracted.”
How can we use EEG and MEG?
Spatial resolution of MEG and EEG is
much worse than fMRI
* Localizing the spatial source of the
signal is challenging, because the
responses detected at the scalp contain
a mixture of signals from many parts of
the brain
* Known as the ”source-localization
problem”
