Scribd
Upload
104 views26 pages

Intro To Neurons and Eegs Brendan Allison, Ph. D. Bci Class Lecture #2

This document provides an introduction to neurons, electroencephalography (EEG), and how EEG is used to study brain activity. It discusses that neurons communicate via electrical signals, and when millions fire simultaneously it produces detectable electrical activity on the scalp. EEG measures this brainwave activity using electrodes placed on the head. Common uses of EEG include studying event-related potentials to external stimuli, spontaneous brainwave patterns, and applications in medical diagnosis, research, entertainment, and monitoring brain states.

Uploaded by

Rajesh Chaurasiya
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
104 views26 pages

Intro To Neurons and Eegs Brendan Allison, Ph. D. Bci Class Lecture #2

This document provides an introduction to neurons, electroencephalography (EEG), and how EEG is used to study brain activity. It discusses that neurons communicate via electrical signals, and when millions fire simultaneously it produces detectable electrical activity on the scalp. EEG measures this brainwave activity using electrodes placed on the head. Common uses of EEG include studying event-related potentials to external stimuli, spontaneous brainwave patterns, and applications in medical diagnosis, research, entertainment, and monitoring brain states.

Uploaded by

Rajesh Chaurasiya
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 26

Intro to neurons and EEGs

Brendan Allison, Ph. D.


BCI class lecture #2
These are neurons. Your brain has hundreds of
billions of them!

Diagram of a neuron. A group of real neurons.


Brainwaves (EEGs) reflect the brain’s electrical activity. A
neuron at rest is like a little battery. Whenever a neuron is
active, its voltage briefly changes. If millions of neurons all
fire at the same time, this produces electrical activity
detectable to an electrode placed on the head.

Diagram of a neuron. Your brain When a neuron is active, its voltage may
has hundreds of billions of them! change by 100 mV or more
When millions of neurons fire at the same time, they may
produce electrical activity detectable to an electrode placed
on the head.

Two real human brains. The left image is an MRI.

Two illustrations of the brain producing electricity.


For example, if you hear a tone, many different groups of
neurons activate to process that tone. EEGs can tell us when
and where these groups of neurons fire. Doctors often use
this technique to diagnose hearing disabilities, since EEGs
can reveal which groups of neurons are damaged.

This figure shows some of the


EEGs evoked by a tone. Early
responses (within 100 ms of the
tone) are very consistent. Later
EEG components may vary
depending on whether you
ignored the tone, if it was
meaningful to you, if you
expected it, and other factors.
Macroelectrodes only measure the coordinated activity of
many millions of neurons. Microelectrodes only measure
the activity of one or very few neurons.

The most common recording setup is a scalp macroelectrode.


While it is possible to get data from as few as two
electrodes, most labs use an electrode cap. These caps are
specially designed so that each electrode is over a general
region of the brain. This makes it easier to estimate the
source of any EEG activity detected at each electrode.
Most EEG recordings use an electrode cap that contains a
large number of electrodes. Many labs use between 16
and 64 electrodes, but caps with 256 or more electrodes
have been used in scientific and medical studies.
Most electrode caps are designed with electrodes over
specific areas of the skull (and thus specific areas of the
brain). Otherwise, you would be recording from different
brain areas each time you use a cap.

These are standardized electrode locations, called the International 10-20 system.
Limitations of macroelectrode recording from the scalp:
1) Difficult to know exactly what brain region is
underneath each electrode
2) Scalp smears electrical signal
3) Scalp acts as low pass filter
4) Only measures neurons near the scalp
5) Only measures neurons perpendicular to scalp
6) Neurons aligned opposite each other cancel each
other
7) Neurons must be active in synchrony to be detected
Unlike macroelectrodes, microelectrodes can only record
brain activity if placed in or on the brain. This is only
done when medically necessary. Typical reasons
include finding the source of seizures or preparing to
implant a device into the brain, such a deep brain
stimulation device used to treat Parkinson’s disease.

The Utah Intracranial Electrode Array The cone electrode


Scalp macroelectrode Implanted microelectrode

Safe and easy Requires neurosurgery


Requires minimal training Requires neurosurgeon
Measures combined activity Measures activity of
of many millions of neurons one of very few neurons
Poor localization Excellent localization
Signal is degraded as it Signal not degraded; can learn
travels through the skull more details of neural activity
Susceptible to noise Less vulnerable to noise
Chronic use definitely OK Chronic use probably OK
1. It is often necessary to place an electrode on or behind the
ear before donning the electrode cap. Scientists often
clean the area behind the ear with rubbing alcohol. Some
people put electrodes near the eye to detect blinking and
other eye movements.

2. The scientist measures the subject’s head and then places


the correct sized cap on his head.
3. Electrode gel is then placed between each electrode and
the scalp to get a good connection. Everyone agrees that
electrode gel in your hair is a wonderful experience.

Two types of electrode gel.


Squirting gel under an electrode.
4. The scientist checks the cap to make sure there is a good
connection between each electrode and the brain.

5. The subject is now ready for recording! A typical


recording session lasts about an hour. It takes roughly 30
minutes to prepare a subject for recording, depending on
the number of electrodes, the subject’s hair, the scientist’s
skill, type of electrode cap, and other factors.
6. After recording, the cap is removed. Electrode gel washes
out easily with water, so many subjects rinse or wash their
hair after a recording session. Of course, smart people
know that electrode gel in your hair makes you cool.

7. That’s it! The subject is done, but the scientist now has
data to analyze.
Event related potentials (ERPs): Brain’s response to a specific
event, such as a tone or flash.
Spontaneous or free-running EEG: Naturally produced,
rhythmic brainwaves; do not require outside activity.

Commonly studied ERPs include the P300 and N400.


Well known free running EEGs include:
Delta (1-4 Hz), found in deep sleep
Theta (4-8 Hz), found in sleep, meditation, hypnosis
Alpha (8-14 Hz), indicate relaxation and closed eyes
Mu (8-14 Hz), largest when individual is not moving
Beta (non specific higher frequencies), indicate alertness
This graph shows about four seconds of EEG from a human subject. Each of the 15 lines
represents a different electrode site. This has a lot of alpha activity (about 10 waves per
second), meaning the subject was probably awake but drowsy with eyes closed.
Remember that alpha waves are a type of free running EEG
These two figures show responses to flashes. Specifically, they
show how the brain responds differently to flashes people
notice compared to flashes they choose to ignore. In each
graph, the relatively flat lines (red or dashed) show the brain’s
response to ignored flashes. Notice the large bump starting
around 300 milliseconds in the other lines (blue or solid),
showing the response to flashes people counted. These are
examples of the P300, a type of Event Related Potential (ERP).

Top figure: from Allison


thesis (Allison 2003)
Bottom figure: from Bayliss
thesis (Bayliss 2001)
Filtering: This removes unwanted portions of the EEG.
There are 4 types of filters: highpass, lowpass, notch,
bandpass. Common settings include .1-100 Hz, notch at
60 and 72 Hz.
Amplification: Data are typically amplified 10,000 times.
Digitization (A/D conversion): The analog signals from the
brain must be translated into digital signals. Filters are
often set at 8, 12 bit or 16 bit. 8 bits = 0-256.
Epoching: ERP data must be aligned with external events.
There are many techniques for studying the brain. Many of
these, such as X-rays, CAT scans, or a standard MRI,
only measure brain structure. They do not measure brain
function. That is, they give you the same picture no
matter what you are thinking or doing. Thus, they would
not be useful in BCIs, because BCIs must be able to
distinguish different brain states.

Some people use EEGs in combination with fMRI. This can


be a very powerful tool for finding exactly when and
where something occurs.
Here are three other technologies for studying brain function:
Functional MRI: These measure blood flow. The subject
must be placed inside a huge magnet. Different brain
areas give a different magnetic signal depending on how
much blood they are using.
PET scans: These measure radioactive decay. Radioactive
material is injected into the subject’s bloodstream, and
thus areas that use more blood emit more radiation.
Magnetoencephalogram (MEG): This measures the brain’s
magnetic activity, not electrical activity. It provides
somewhat different information than EEGs.
Why haven’t these technologies been used for BCIs?

First, all of these require millions of dollars of nonportable


equipment, and a highly trained technician.
Second, PET and fMRI measure blood flow. When you start
using a brain region more than usual, it takes a few
seconds for blood flow to increase. Thus, only EEG and
MEG can measure brain activity in true realtime.

There are other reasons too; these are the main ones.
Pure research: EEGs help us learn more
about when, where, and how different brain
areas work together when thinking, speaking,
responding to tones, etc.
Medical: Isolate areas or processes that
respond slowly, improperly, or not at all.
Detect onset of seizures, strokes, psychotic
episodes, or other problems. Enable
communication for severely disabled with
brain computer interface systems. Train kids
with ADD to pay attention longer. Study the
effects of drugs over time.
Entertainment/relaxation: Some people
have used EEG systems designed for alpha
biofeedback. This means that you train
yourself to have more alpha waves in your
EEG, which helps some people relax. EEGs
can be used to play simple games or make
music.
Passive monitoring: There has been a lot of
progress recently in alertness monitors based
on the EEG. These might warn people they
are dozing off. This has been proposed for
people in “attention-critical” situations like
nuclear plant technicians, sonar operators,
security guards, and truck drivers. Lie
detection may be possible too.

You might also like