What's amazing about this playlist is not just that it is free for everybody to see; it's also so easy to follow. This professor is so good at explaining these complex topics that even me, a complete layman, (I am a software engineer) can easily follow. It's sixth lecture and I had to stop only couple of times too google some terms (for example here I didn't know what BOLD response is). It's doing a great service to those of us that are interested in cognitive science, but don't have time to enroll on a course or read those tiresome scientific papers lol
I have been watching your lectures every night after the dinner for about a month now. It's been super interesting ride and I like how the professor has a dignity and respect to her field and people in it. How she talks and teaches not only encourages people to study and learn but also makes people ask more questions in their daily lives and enrich their lives and seek for more. I hope she hears how much people appreciates her lectures!! (I mean she probably does hahaha) Thanks for people involved in this video for doing this too!
I think I understand what the instructor meant about using VGG-face to distinguish between images of chairs and cars. The model's last layer before producing a categorical output generates embeddings (high-dimensional vectors). What they likely did-or what I would do-is average these embeddings to create a single vector for each image, then cluster all of the vectors from all images into groups. If the model could effectively differentiate between chairs and cars, you’d expect to see two clear clusters when visualizing the embeddings. This would confirm the model’s ability to distinguish the two categories. However, I suspect the clusters weren’t perfectly distinct but still showed signs of forming two groups. I’m curious if they tested this approach with more object categories.
Professor, these videos are phenomenal, you are a fantastic lecturer. Do you have an opinion about the mind vs. the brain. Is our mind only in our brain or does our mind extend beyond our physical being?
I don't get the argument by Haxby. Images of faces are not faces too! FFA doesn't tell what is face and what is not face, it tells how similar to the face the thing which our eyes sees. And obviously different object have different amount of similarities to faces. Thereby the thing which is created to distinguish faces is EXPECTED to allow us to distinguish most of the other objects too.
don't give up your day job... even a dcnn can tell you a picture of a face casts an image on its retina just as a real face does, which is why it and we can recognise faces in pieces of toast and clouds and the moon and etc
I am curious how to scan a monkey in MRI and make sure the data are of good quality. What if the monkey moves its tongue or wouldn’t stop moving its head? Those can also be alternative explanations why MRI data do not work to the level of neuron recording. They may also use different decoding algorithms. I hope there are still potentials to try more ways.
Any part of this module discussed about "addiction". What is the process in the brain that drives us to addiction of something (game, music, drug, pornography, etc.)? I would love to hear science behind it.
The man with the electrodes around the ffa area knew that he saw face patterns on objects, that could imply that there are still selective neurons or voxels for face patterns right? What if you measure these voxel patterns of that area like Haxby did for objects and compare those to faces. I would speculate that you first need to identify the object before seeing the face so you have to distinguish a little bit. If I try, I see faces everywhere on objects etc but I don’t see that by default. (This is my first toe dip into the human brain so I lack some knowledge)
I didn't quite get the FFA response interaction estimates in 31:35 (All of the 4 stimulus effect conditions) Can someone explain these estimates that the volunteers made? P.S. I haven't framed my question right, but I hope you get a clue to what I mean
Hi, let me try: - the first student was given the task to show an example of measurements in the case where: a) the main effect wasn't present b) whether or not subjects were attentive also didn't influence the results c) there was no interaction between a) and b) She therefore drew two arbitrary dots, where the only important thing was that they are of the same height, meaning the measurement was the same. Using the example they explored, this meant that subjects didn't select for faces any more than for objects AND that this measurement didn't change when the attention variable was introduced AND that there was not even a linear difference between attentive and non-attentive results. - the second student was tasked with showing measurements where: a) the main effect WAS present b) attention remained a non-factor c) still no interaction between a) and b) This meant that there was a differece in height of the two dots, meaning that subjects selected for faces more frequently (or quickly) than for objects, BUT it also meant that the attention factor didn't influence these measurements NOR was there a linear difference between attentive and non-attentive tests. - the third student had the task where a) the main effect was present b) attention WAS a factor c) still no interaction between a) and b) This meant that, as with the previous example, subjects did select more for faces than objects AND this time the attention factor influenced the amount (or quickness) of selection, BUT this attention effect was linear relevant to both tasks. In other words, the subjects' results for both faces and objects was influenced by inattention equally. - finally, the last student was tasked with showing: a) main effect present b) attention being a factor c) an interaction between a) and b) This meant that, similar to the previous example, the subjects still selected more for faces than objects, the attention factor still resulted with worse results AND the change in results due to attention factor was more pronounced for objects (meaning subjects, when not paying attention, were exponentially worse in selecting for objects than for faces). Now, it's important to say that the change shown by the fourth student is just one possible change that could have happened. But this was a valid one, after a few corrections by the professor.
Great lectures. Maybe humans spend much more time recognizing faces and reading expressions, which could contribute to the evolution of the FFA region. It would be interesting to examine other species, such as predators, and herbivores, they may pay more attention to the gesture of their rivals to survive (also rely more on auditory), rather than Face. Or species with no face, like the faceless void.
What I don't understand is that a picture of a face is just a picture and not a face. Looking at a picture is not like looking at real face....see what I mean?
I wonder if some types of objects or other symbols would earn a place in one's brain, if they were closely and consistently enough present in one's live as important part of it. Say, like paintings to a painter, or some types of tools to an artisan.
Humility is important ...my father says without humility...I think this whole looking at people as patients is backfiring. Differently abled. Trouble fitting in. Need time to learn. Maybe diagnosis after 25.? Or write as unknown Comfort with not knowing a prerequisite for science?
Face feels a bit more same....bodies so many differnce....we have face affected by various things..like my mouth didn't have braces... Sleep apnea....worse brain? Like pregnant woman here take a lot of rest..... considered..good due to sicence...which changed to can take moderate walks...etc.
I'm trying to get my parents to touch me... refrigerator like...weird..no touching from mom...... Dad does this weird thing.....like a robot brushes hair robotically. Think mom thinks she is a genius, who thinks, it's a waste of time of her iq .. Something like that.
What's amazing about this playlist is not just that it is free for everybody to see; it's also so easy to follow. This professor is so good at explaining these complex topics that even me, a complete layman, (I am a software engineer) can easily follow. It's sixth lecture and I had to stop only couple of times too google some terms (for example here I didn't know what BOLD response is). It's doing a great service to those of us that are interested in cognitive science, but don't have time to enroll on a course or read those tiresome scientific papers lol
I have been watching your lectures every night after the dinner for about a month now. It's been super interesting ride and I like how the professor has a dignity and respect to her field and people in it. How she talks and teaches not only encourages people to study and learn but also makes people ask more questions in their daily lives and enrich their lives and seek for more. I hope she hears how much people appreciates her lectures!! (I mean she probably does hahaha) Thanks for people involved in this video for doing this too!
Feeling GRATEFUL for having this lecture series at the tips and taps of my right thumb, too cool!
Lecturer has a great attitude to being a researcher, it comes through.
informing and entertaining at the same time, amazing lecturer
Thank you so very much Nancy Kanwisher, MIT opencourseware, UA-cam and all I'm missing very educational and easy to follow
Helps me feel like I am having some meaningful time whilst I am at home struggling with depression and motherhood.
Studying science is my way of chasing dopamine
Lecture 7. I think I’m falling in love
I think I understand what the instructor meant about using VGG-face to distinguish between images of chairs and cars.
The model's last layer before producing a categorical output generates embeddings (high-dimensional vectors). What they likely did-or what I would do-is average these embeddings to create a single vector for each image, then cluster all of the vectors from all images into groups.
If the model could effectively differentiate between chairs and cars, you’d expect to see two clear clusters when visualizing the embeddings. This would confirm the model’s ability to distinguish the two categories.
However, I suspect the clusters weren’t perfectly distinct but still showed signs of forming two groups. I’m curious if they tested this approach with more object categories.
23:05
Is nice inspiring definition
I'll die for her lol...... sooooo good. And it so opening up my mind, an Indonesian citizen
Professor, these videos are phenomenal, you are a fantastic lecturer. Do you have an opinion about the mind vs. the brain. Is our mind only in our brain or does our mind extend beyond our physical being?
Her lectures are super interesting!!
I don't get the argument by Haxby. Images of faces are not faces too! FFA doesn't tell what is face and what is not face, it tells how similar to the face the thing which our eyes sees. And obviously different object have different amount of similarities to faces. Thereby the thing which is created to distinguish faces is EXPECTED to allow us to distinguish most of the other objects too.
don't give up your day job... even a dcnn can tell you a picture of a face casts an image on its retina just as a real face does, which is why it and we can recognise faces in pieces of toast and clouds and the moon and etc
I am curious how to scan a monkey in MRI and make sure the data are of good quality. What if the monkey moves its tongue or wouldn’t stop moving its head? Those can also be alternative explanations why MRI data do not work to the level of neuron recording. They may also use different decoding algorithms. I hope there are still potentials to try more ways.
Thanks 🤍❤️
Any part of this module discussed about "addiction". What is the process in the brain that drives us to addiction of something (game, music, drug, pornography, etc.)? I would love to hear science behind it.
The man with the electrodes around the ffa area knew that he saw face patterns on objects, that could imply that there are still selective neurons or voxels for face patterns right? What if you measure these voxel patterns of that area like Haxby did for objects and compare those to faces. I would speculate that you first need to identify the object before seeing the face so you have to distinguish a little bit. If I try, I see faces everywhere on objects etc but I don’t see that by default. (This is my first toe dip into the human brain so I lack some knowledge)
I didn't quite get the FFA response interaction estimates in 31:35 (All of the 4 stimulus effect conditions) Can someone explain these estimates that the volunteers made?
P.S. I haven't framed my question right, but I hope you get a clue to what I mean
Hi, let me try:
- the first student was given the task to show an example of measurements in the case where:
a) the main effect wasn't present
b) whether or not subjects were attentive also didn't influence the results
c) there was no interaction between a) and b)
She therefore drew two arbitrary dots, where the only important thing was that they are of the same height, meaning the measurement was the same. Using the example they explored, this meant that subjects didn't select for faces any more than for objects AND that this measurement didn't change when the attention variable was introduced AND that there was not even a linear difference between attentive and non-attentive results.
- the second student was tasked with showing measurements where:
a) the main effect WAS present
b) attention remained a non-factor
c) still no interaction between a) and b)
This meant that there was a differece in height of the two dots, meaning that subjects selected for faces more frequently (or quickly) than for objects, BUT it also meant that the attention factor didn't influence these measurements NOR was there a linear difference between attentive and non-attentive tests.
- the third student had the task where
a) the main effect was present
b) attention WAS a factor
c) still no interaction between a) and b)
This meant that, as with the previous example, subjects did select more for faces than objects AND this time the attention factor influenced the amount (or quickness) of selection, BUT this attention effect was linear relevant to both tasks. In other words, the subjects' results for both faces and objects was influenced by inattention equally.
- finally, the last student was tasked with showing:
a) main effect present
b) attention being a factor
c) an interaction between a) and b)
This meant that, similar to the previous example, the subjects still selected more for faces than objects, the attention factor still resulted with worse results AND the change in results due to attention factor was more pronounced for objects (meaning subjects, when not paying attention, were exponentially worse in selecting for objects than for faces). Now, it's important to say that the change shown by the fourth student is just one possible change that could have happened. But this was a valid one, after a few corrections by the professor.
@@oNtuobAwoH thanks for taking the time to clarify that.
@@oNtuobAwoH Further clarification for @Et Cetera:
Attended= Attention given
Unattended= Attention not given
Main effect= Stimulus
Is the reading list available for the course?
ocw.mit.edu/courses/9-13-the-human-brain-spring-2019/pages/readings/
Best wishes on your studies!
This is helpful 🤍❤️
Great lectures. Maybe humans spend much more time recognizing faces and reading expressions, which could contribute to the evolution of the FFA region. It would be interesting to examine other species, such as predators, and herbivores, they may pay more attention to the gesture of their rivals to survive (also rely more on auditory), rather than Face. Or species with no face, like the faceless void.
Also deep face recognition requires more energy than recognizing other objects...maybe
What I don't understand is that a picture of a face is just a picture and not a face.
Looking at a picture is not like looking at real face....see what I mean?
I wonder if some types of objects or other symbols would earn a place in one's brain, if they were closely and consistently enough present in one's live as important part of it. Say, like paintings to a painter, or some types of tools to an artisan.
Great lecture
Nice lecture.. thanks MIT
No region for flügelhorns? That can't be right
Nice work
Can we use combined data from animals(rat,dog,monkey).
On account of humans similar in different areas?
Took an oleanz 2.5 and mitrazapine 3 mgs.
No sadness.... tranquilizer.
is it carrie ? 😆
Hi everyone!
It's easier on a rectangular paper?
Humility is important ...my father says without humility...I think this whole looking at people as patients is backfiring.
Differently abled.
Trouble fitting in.
Need time to learn.
Maybe diagnosis after 25.? Or write as unknown
Comfort with not knowing a prerequisite for science?
Anyone else watching this for the first time excited for Neuralink?
16:50 🐙🤣 make a new captcha
I would love to be scanned while multitasking during an audio lecture 🥰 art and Feynman....guitar and sermon
Nice
Math ability changes with personality?... environment?
too late
😮
I am analysing each word?
Cofffeeeeeeeee😢
Water....all these factors
The brains IPA..
Blocks SERT and SNRI epinephrine.
Psychology...the science of gaps... temporary fix
....before we know the whole story.
You are my favorite women
Brain leaves challenges. During lecturea..it wlll digest.
Face feels a bit more same....bodies so many differnce....we have face affected by various things..like my mouth didn't have braces...
Sleep apnea....worse brain?
Like pregnant woman here take a lot of rest..... considered..good due to sicence...which changed to can take moderate walks...etc.
I think I hold ppl/neuroscients to a pedestal?
Writing....physics...biology
For ill stepping up slowly good. Parents said this...big believer in dnaism.
Cold/hot.....all factors have to be taken of the patient.
Fill in these factors.
Temp
What u like.
Food.
Sex...done or not.
House
Patients read journals?
Summaries.
We wanted to be British when British wanted to be indian.....both saw science in other.
Mittens school
I'm trying to get my parents to touch me... refrigerator like...weird..no touching from mom......
Dad does this weird thing.....like a robot brushes hair robotically.
Think mom thinks she is a genius, who thinks, it's a waste of time of her iq ..
Something like that.
Go easy on movies kids...👳
U realize u r being therapized? U r like thinking why my brain entering?
"Now i will be entering the brain"
"Is that ok"..."it's not chemical"
Mit is so tensed up....cultural fests....too much intellectual ization
What are you talking about? Educating people on advanced math and science is exactly what people should do. There's nothing bad about it.
Boring af topic. But I have crush on the proff
GREAT EXCELLENT A GENJUUS IS DISCHOVERED
-- GOOTO FURH DANMARK ET MOI - KUMKAMKARTEL -CHICAGO -ONE EMPLOYEE LMFAO SEALCLATm