SMV HUMAN NEUROPHYSIOLOGY
The Resting state: physiology, cognition and your projects
Cognitive neuroscience is a combined study of mind and brain. Non-invasive neuroimaging
techniques play an important role in furthering this field. This allows exploring the biology of:
conscious experience, unconscious processes and the disordered mind. This aims to
improve diagnosis and treatment of brain-disorders and increasingly inspires artificial
intelligence. These are brain inspired algorithms.
In the early days PET imaging played an important role. Here radioactive materia (FDG)l is
injected. FDG sugar accumulates in brain areas that are metabolically active. The decay of
FDG gives off a neutron and positron, the positron hits an electron and annihilates into 2
gamma rays which are detected. PET has given us insight into the metabolism of the brain.
The temporal resolution is 10-20s, this is not so great, and the spatial resolution is 5mm.
fMRI has replaced PET in cognitive neuroscience, this is how fMRI works. The brain
consumes about 20% of the energy of the whole body. The brain needs oxygen for that, it
dilates the capillaries for that. In the regions of the brain where much energy is consumed
there is a high concentration of oxygenated hemoglobin. De- and oxygenated hemoglobin
have different magnetic properties. You can measure the magnetisation of brain tissue from
second to second with the fMRI, if you measure change you can see that there has been
activity there. This is a secondary effect to neuronal activity.
You have to take functional specialisation of the brain regions with a grain of salt. They are a
bit specialized but a lot of the time the whole brain is still involved, some regions more than
others. You also have to think about what you’re comparing. You
can’t simply subtract the passive from the active state, because no
passive state exists where there is no passive state with little to no
brain activity. Sometimes in an active state the brain activity in
certain regions is even less than in the passive state.
This gives an idea that there is a default mode of the brain that is
active during rest that changes during goal-oriented behaviour. The
regions that are deactivated during this behaviour are also
functionally connected, they are in synchrony and communicate. It is
thought that the default mode is due to a lot of daydreaming. This
leads to people feeling less happy.
In a lab setting people dit an eyes closed rest experiment and then
people filled in a questionnaire, ARSQ. This gives us insight into 10
dimensions and couples neuroimaging with cognition.
,Neuronal basis of EEG and ERPs
Electroencephalography (EEG)
We research with EEG because it tells us something about neuronal activity and gives us
insight into some interesting questions. There are a few advantages over PET and fMRI:
direct reflection of activity, high spatial and temporal resolution (1 ms), greater specificity,
increasingly portable, non-invasive, more available and inexpensive.
This is how an EEG recording goes. We have to see neurons as little batteries: the
concentration of ions is not the same on the inside and outside of the cell. When a
neurotransmitter causes the opening of a transmembrane ion channel, a current of ions will
flow from the outside to the inside of the cell. That causes a transient potential
around the neuron, and because the head is full of water, which conducts
electric fields, electric potentials are propagated to the outside of the head.
These potentials are different in different places on the head, because the field
decays with distance to the electric-dipole source. Thus, by measuring the
potential difference with electrodes having good electrical contact with the
scalp, we can measure brain activity. Only, because of each neuron producing
a very small electric dipole (separation of positive and negative charge) many
neurons need to be active, which can happen either because of some
stimulus/task event, or spontaneously. The lettering of the electrodes refers to
the main underlying anatomical lobes.
We always measure potential differences with the electrodes. This means we
always need to have a reference electrode. Cz is the common reference electrode, you
measure the difference between the Cz and all the other electrodes. But we can’t reference
Cz to itself, we have to make a new reference: the average reference. This is the average of
all electrodes and can be positive or negative. We want to do this because otherwise it
seems like the activity at Cz is 0.
We have different factors influencing the EEG signal that we’re analyzing:
1. Position of the reference electrode. You can move it from Cz to average, for eg.
2. Electrical contact between scalp and electrode. Air is not a good conductor.
3. Conductivity of the head. Bone is not a good conductor, a thick or thin skull will have
influence on the signal.
4. Number of active neurons.
5. Synchronicity between the neurons. The more, the larger the activity on the scalp.
6. Orientation of the cells. We have star-shaped interneurons and parallel cells. The
star-shaped has a closed field configuration which does not create an electric field at
long distances. The parallel pyramidal cells support spatio-temporal summation of
currents.
7. Distance from source to electrode. Electric field decays with distance, so the EEG
measures cortex activity the most and deeper brain region activity the least.
8. Orientation of the diple. The same activation on opposite sides of a sulcus will
change the polarity and the signal is only maximal right above the source if the dipole
is radial. Different people can have opposite polarities for specific components. The
anatomical location of the source generating the EEG signal is not easy to deduce
from its corresponding scalp topography.
, 9. Site of activation on the neuron. Superficial and deep layer activation may lead to
opposite EEG polarity.
10. Excitation vs. inhibition. Excitation is active sink passive source, and inhibition is
active source passive sink.
11. Artifacts: EEG signals that don’t come from brain sources.
EEG vs. MEG
MEG is based on magnothermaters that need to be cooled by liquid helium. It measures
scalp potential differences, the absolute field, and it has single sensor activators. Magnetic
fields are created by the currents and go through the skull. They are difficult to stop, hardly
deflected. The magnetic barriers of physiological tissues are almost not there, not a lot of
things affect the MEG. This causes the spatial resolution to be better than EEG. It is more
expensive and not portable though.
The event-related potential technique
There are two main categories of EEG: event-related response (ERP) by a task or stimulus
and spontaneous activity. An evoked response/ERP is a transient deflection in EEG/MEG
signals caused by a stimulus, cognitive task or motor activity. Usually an average of many
trials is used and baseline corrected. This means that the signal in a pre-stimulus interval of
50-100 ms is defined to be 0.
A lot of the brain activity is spontaneous approx 95%, only 5% is related to processing
sensory stimuli. When you activate this type of process in the brain it comes on the top of the
spontaneous stimuli. The response of one person is not a very clear signal, this means you
have to average out your signals to see an evoked response. Low signal to noise ratio
means that there are more trials necessary. You can see however that there is a very clar
dipolar field when you look at the scalp deflection in an ERP.
