Lecture 1 - Brain structure, anatomy, APS

Question 1

Neurosicence

Answer 1

study that aims to understand the human brain and how it works.

Question 2

AI

Answer 2

Artificial systems that mirror human intelligence and are able to take over tasks that require human intelligence.
can be divided in: (1) Narrow AI( ANI ), (2) General AI( AGI ) and super AI( ASI ).

Question 3

AI benefits from Neuroscience

Answer 3

-Robustness of inference: inference about data that is insensitive to smaller or larger deviations from the assumptions under which is derived.
Neuroscience can make AI more robust, adaptive and efficient (e.g. image recognition).

• Data efficiency: efficient handling of data, through the means of transfer learning (building on previous knowledge), unsupervised learning and evolutionary constraints.
• Energy efficiency: efficient handling of energy, as the brain is much more efficient with energy, than any AI system we currently have.
• Existence proof and inspiration: The fact that the brain is the only existing proof of general intelligence.
• Human machine interaction: A study that aims to improve interactions between humans and computers/machines. observing human behaviour could help create intelligent systems able to replicate such behaviour.

Question 4

Neuroscience benefits from AI

Answer 4

-Finding patterns in hidh-dimensional data: patterns are series of data that repeat in a recognizable way,
For high-D data, today’s computational neuroscience data would defenitly need machine learning techniques from AI to recognize such patterns.

• Task-performing computational models: Deep neural networks provide the best explanation of computations in several human systems, such as the visual system.
• Dealing with big data: AI can help pre-process large amounts of data, integrate multiple signal sources and perform complex statistical modelling on such data.

Question 5

Cerebrum

Answer 5

The largest part of the brain, located either on top or in front of the brainstem.
It contains the cerebral cortex as well as many other structures.

Question 6

Cerebellum

Answer 6

Smaller part of the brain, mainly responsible for motor control

Question 7

Cerebral cortex

Answer 7

The outer layer of the neural tissue of the cerebrum.
The cerebral cortex can be divided into multiple areas naimly the four cerebral lobes: (1) frontal lobes (the front part), (2) parietal lobe (middle top part), (3) occipital lobe (back part), (4) temporal lobe (side part)

Question 8

Slicing through the brain

Answer 8

Axial- Anterior - posterior axis & laterla - medial axis.
Coronal: Dorsal - ventral axis & laterla - medial axis
Sagittal: Anterior - posterior axis & Dorsal - ventral axis.

Question 9

The folded appearance of the brain (cotrical folding)

Answer 9

Is caused by:
-Gyri ( s: gyrus): ridges on the cerebral, generally surronded by one or more sulci.

Sulci ( s: sulcus): grooves in the cerebral cortex that surrond gyri.
Large sulci are called fissures.

With an increase in cognitive functioning, the brain surface area expands. To account for this expansion with the restricted space in the skull, the brain surface area is folded.

Question 10

Cortical layering

Answer 10

The cerebral cortex mostly consists of the six-layered neocortex, which is involved in high-order brain functions such as sensory perception and motor commands, each layer contains a characteristic distribution of different neurons.

Question 11

Neurons

Answer 11

Electrically excitable cells that communicate with other cells through synapses. They consist of a cell body (soma), several dendrites (recieve signal) and a single axon (transport signal).
-learning is implemented by changing the synaptic connectivity between neurons.

Question 12

Action potential

Answer 12

An explosion of electrical activity, also called an impulse.

Action potentials are binary as the only ‘fire’ or don’t ‘fire’ depending on whether the threshold70 mV is reached.

Question 13

Membrane potential

Answer 13

The difference in electric potential between interior and the exterior of a cell.
Membrane potentials are continious\graded as they depend on the permeability of ion channels (the gates between interior and exterior) and the difference in ion concentrations between exterior and interior.
-post-synaptic potentials are chnages in the membrane potential that either initiate potentials (excitator PSP) or inhibit them (inhibitory PSP).
They are monophasic and slow.

Question 14

Neural firing

Answer 14

is often accompanied by a lot of (stochastic) noise.

Question 15

Neural coding

Answer 15

code consisting of neural responses at different timescales (how often or fast they fire in a specific timeframe) tp represent sensory information.

Question 16

Time relationship (action potential timing)

Answer 16

between on neuron firing and another neuron firing, also contains part of the sensory information. distorting this will result in a loss of information.

Question 17

Neurons fire together in harmony

Answer 17

can be measured by an EEG.

An EEG will provide a neural oscillation in which you can clearl see such a rythm.

Question 18

Brain size vs intelligence

Answer 18

The idea that humans are special because of their intelligence, can not be explained by the size of our brain or total amount of neurons in comparison to other animals (such as elephants).
It is likely because we have a lot of neurons in the cerebrum of the brain, while most of the neurons in elephants are in their cerebellum.

within species, the size and number of neurons scale linearly (at different rates). Rodents for example, have a different scaling rate than primates for different sizes and numbers of neurons.
- we have as many neurons as expected from a large primate.

There is no difference between male and female when it comes to IQ, despite males having a larger brain size.
Within males and females respectively, larger brains would likely lead to higher iq’s but but between them, this is defenitly not the case.
Since females have smaller brain sizes than males, it can be stated that female brains work more efficiently.

There is a trade-off across species between the robustness of neural code and it’s efficiency,
Human neural code is less synchronous (less robust),
but show higher information capacity (increased efficiency).