EEG LAB ORIENTATION OCTOBER 1, 2004

1
EEG LAB ORIENTATIONOCTOBER 1, 2004

• JACK GELFAND
• RAFAEL ESCOBEDO
• COGNITIVE ELECTROPHYSIOLOGY
• LABORATORY
• jjg_at_princeton.edu
• http//csbmb.princeton.edu/eeglab

2
INTRODUCTION TO EEG

• Origin of EEG signals
• EEG techniques
• Properties of EEG signals
• Event-related averaging
• Time frequency analysis
• Source localization

3
Electroencephalography - the recording of
electric currents generated in the brain, by
means of electrodes applied to the scalp, to the
surface of the brain (intracranial), or placed
within the substance of the brain
(depth)Electroencephalogram - a recording of
the potentials on the scalp generated by currents
emanating from the nerve cells in the brain
Electroencephalograph - an instrument for
performing electroencephalography
4
HANS BERGER WAS THE FIRST TO REPORT ELECTRICAL
SIGNALS RECORDED FROM THE SCALP OF HUMAN SUBJECTS
Berger, H., On the Electroencephalogram of Man,
1929.
Berger recorded alpha and beta waves and
demonstrated that they changed with mental state.
5
ORIGIN OF THE EEG SIGNAL

• The electrical current flowing in adjacent brain
  tissue produced by the firing of a single neuron
  is very small.
• When a large population of neurons are active
  together, they produce electrical currents large
  enough to be detected by electrodes placed on the
  scalp.
• Changes in potential between a recording
  electrode and a reference electrode can be
  measured as a result of this current flow.
• This methodology is known as electroencephalograpy
  (EEG)
• The record of the signals is referred as the
  electroencephalogram.

6
CORTICAL PYRAMIDAL CELLS PRODUCE OBSERVABLE
ELECTRICAL SIGNALS
7
POLARITY DEPENDS UPON POSITION OF SYNAPSE
8
EEG AND MEG HAVE BETTER TIME RESOLUTION THAN fMRI
BUT WORSE SPATIAL RESOLUTION
The time resolution of electrophysiological
techniques allows you to resolve individual
frequency components of the waveform. These
frequency components have a direct connection to
brain mechanisms.
9
HOW DO WE MEASURE AND RECORD EEG SIGNALS ?

• Measure potentials generated by current flow
• Signals are 1 -100 microvolts
• Analog signals must be digitized and saved.
• Data acquisition must be synchronized with the
  stimulus presentation process

10
EEG LAB CONTROL ROOM
11
SUBJECT TESTING ROOM
12
EEG CAP
13
EEG RECORDING HARDWARE CONFIGURATION
COGNISCAN SYSTEM
The system is designed around two National
Instruments 6071E 64 channel analog to digital
converter cards and is controlled by LabView
software.
14
COGNITIVE ELECTROPHYSIOLOGY LAB CAPABILITIES

• 128 Channel EEG, 4 Channel EMG
• E-Prime text, picture and sound stimuli
• DVD movie with event markers
• Matlab and Presentation compatible
• Hidden camera for subject observation
• Pupillometry and EEG

15
CORTICAL CONFIGURATION
16
A TYPICAL RESEARCH CAP CONFIGURATIONHAS 64-256
ELECTRODES
17
EEG SIGNALS CHANGE WITH VARIOUS STATES OF
CONSCIOUSNESS
18
EVENT RELATED POTENTIAL AVERAGING
19
ERP WAVEFORMS ARE ASSOCIATED WITH DIFFERENT
COGNITIVE MECHANISMS
20
DIFFERENT COGNITIVE PROCESSES HAVE DIFFERENT MAPS
OF ELECTRICAL ACTIVITY ON THE SCALP INDICATING
DIFFERENT SOURCES IN THE BRAIN
21
THE ERP IS MADE UP OF MANY UNDERLYING COMPONENTS
Kremlacek et al., Model of Visually Evoked
Cortical Potentials, Physiol. Res.. 51, 65-71
(2002)
22
EEG FREQUENCY SPECTRUM

• Delta - lt4 Hz - deep sleep - very high amplitude
• Theta - 4-8 Hz - drowsiness or light sleep -
  frontal midline theta proportional to mental
  effort
• Alpha - 8-18 Hz - posterior - relaxation, eyes
  closed -Attenuated by attention and mental
  effort.
• Beta - 13 - 30 Hz - anterior - accentuated by
  barbiturates
• Gamma - 30 -100 Hz - local processing

23
TIME-FREQUENCY ANALYSISFrequency
ComponentsPhase Relationship to Stimulus
24
FREQUENCY OF A PURE SINE WAVE
25
COMPLEX FREQUENCY SPECTRA
26
THREE WAYS TO DISPLAY THE FREQUENCY CONTENT OF
EEG SIGNALS
27
WAVELET CONSTRUCTION
28
CREATING A TIME FREQUENCY DIAGRAM
29
(No Transcript)
30
(No Transcript)
31
(No Transcript)
32
ALPHA BAND ANALYSIS IN A SEMANTIC FEATURE
VERIFICATION TASK
stimulus
delay
probe
Experiment 1. Semantic matching task, average of
7 subjects, pseudo-word trials have no task. Note
that alpha desynchronization drop out only
occurs on word trials. No sub-bands are evident
in these figures.
33
ALPHA-BAND SCALP DISTRIBUTION
34
IT IS POSSIBLE TO FIND THE LOCUS OF ELECTRICAL
ACTIVITY BASED UPON THE MAP OF ELECTRICAL
ACTIVITY ON THE SCALP.
35
HEAD MODELS
3-SHELL
The realistic head model uses the head shape from
a structural MRI and known electrical
conductivities of skin, bone, CSF and cortical
tissue.
36
DIPOLE SOURCE MODELING

• There are more unknowns (dipole sources) than
  knowns (electrodes)
• This results in infinite number of possible
  solutions.
• Usually done as an iterative forward algorithm
  with a small number of equivalent source dipoles.
• A starting point is chosen based upon additional
  information -fMRI etc.
• A forward solution of the scalp map is
  calculated.
• An iteration is chosen holding some factors
  constant.
• Orientation
• Location
• Magnitude
• The process is repeated until a satisfactory fit
  of the scalp map is obtained.

37
LORETA - Low Resolution Electro-Magnetic
Tomography

• Overcomes the problem of more unknowns than
  knowns
• Add additional physiologically realistic
  constraints
• Source current density is spatially smooth
• Minimum solution inside the brain - confined to
  surface of cortex
• Finds 3-D distribution of the source-current
  density
• Not a dipole solution
• No assumptions about the number of sources
• Inversion Matrix is calculated once from the
  geometry of the head.

vi
qi
.
.

x
Gij
.
.
ELECTRODEVOLTAGES
CURRENTSOURCES
INVERSIONMATRIX
38
(No Transcript)
39
(No Transcript)
40
(No Transcript)
41
AVAILABLE SOFTWARE IN THE CSBMB EEG LAB

• CogniScan Data Editor
• SOURCE SIGNAL IMAGING
• ELECTROMAGNETIC SOURCE ESTIMATION
• Data Editor
• Source Estimator
• Locator
• MR Viewer
• MatLab Data Analysis
• ERP Averaging
• Wavelet Analysis