eeg

Document Sample
eeg Powered By Docstoc
					The Electroencephalogram
Electroencephalogram (EEG)
 The EEG--an oscillating
  voltage recorded on scalp
  surface
    Reflects Large #
     Neurons
    Is small voltage
 Bands of activity
   Delta 0.5-4 Hz
   Theta 4-8 Hz
   Alpha 8-13 Hz
   Beta 13-30 Hz
   Gamma 30-50 Hz
Utility of EEG

 Relatively noninvasive
 Excellent time resolution
Sources of scalp potentials

 Glial Cells – minimal, some DC steady
 potentials
 Neurons
   Action Potentials – NO, brain tissue has strong
   capacitance effects, acting as Low Pass filter
   Post-synaptic potentials – YES, both inhibitory
   and excitatory from functional synaptic units are
   major contributors
Alpha and Synchronization
 Why Alpha?
   It is obvious and hard to miss!
   Accounts for ~70% of EEG activity in adult human brain
 From where, Alpha?
   Historically, thought to be thalamocortial looping
   Adrian (1935) demolished that theory
       Recorded EEG simultaneously in cortex and thalamus
       Damage to cortex did not disrupt thalamic alpha rhythmicity
       Damage to thalamus DID disrupt cortical alpha rhythmicity
   Thalamic rhythmicity remains even in decorticate preparations
   (Adrian, 1941)
   Removal of ½ thalamus results in ipsilateral loss of cortical
   alpha


                                                          Next
Alpha
Alpha and Synchronization
 Thalamus may drive the alpha rhythmicity of the
 EEG
  Cortex certainly does feedback to thalamus, but thalamus
   is responsible for driving the EEG
  Particularly the reticularis nucleus (Steriade et al. 1985)
 Nunez (1995) has proposed that cortical neurons
 regulate rhythmicity
Recording EEG
Recording EEG
Electrodes, Electrolyte, Preparation

 Reduce impedance
 Ag-AgCl preferred, tin OK
 Electrolyte: ionic,
  conductive
 Affixing
    Subcutaneous needle
    electrodes
    Collodion
    EC-2 paste
    Electrocap
Recording References
 Measure voltage potential differences
   Difference between what and what else?
―Monopolar‖ versus Bipolar
   No truly inactive site, so monopolar is a relative
   term
   Relatively monopolar options
      Body – BAD IDEA
      Head
        • Linked Ears or Mastoids
        • Tip of Nose
   Hypothetical advantages of Monopolar –
   seldom realized
Recording References
 Bipolar recording
  Multiple active sites
      Sensitive to differences between electrodes
      With proper array, sensitive to local fluctuations (e.g.
      spike localization)
 Off-line derivations
   Averaged Mastoids
   Average Reference (of EEG Leads)
      With sufficient # electrodes and surface coverage,
      approximates inactive site (signals cancel out)
      Artifacts ―average in‖
Artifacts
 Three sources
   60-cycle noise
  Muscle artifact
  Eye Movements
Movement in reference lead
Chewing
Vertical eye roll
Excessive muscle – notice saturation of T5
Talking and moving head
Yawn
Eye Closure and reopening
Blink and
Triple blink
Dealing with artifacts
 60-cycle noise
    Ground subject
    60 Hz Notch filter
 Muscle artifact
    No gum!
    Use headrest
    Measure EMG and reject/correct for influence
    Statistically control for EMG
    Hand score
 Eye movements
    Eyes are dipoles
    Reject ocular deflections including blinks
    Computer algorithms for EOG correction
High and low pass filtering
 Do not eliminate
  frequencies of interest
 Polygraphs have broad
  roll-off characteristics
 Digitization rate (Nyquist)

 For example, 0.01 - 100
  Hz bandpass, sampled at
  500 Hz
Time Domain Vs Frequency Domain
Analysis
Time Domain Analysis involves viewing
 the signal as a series of voltages as a
 function of time, [x(0), x(t1), x(t2),...,x(tn-
 1)]
  e.g., skin conductance response, event-related
   potential
   Relevant dependent variables
     latency of a particular response
     amplitude of that response within the time window
 More about time domain in 2 weeks
Time Domain Vs Frequency Domain
Analysis
 Frequency Domain Analysis involves
  characterizing the signal in terms of its
  component frequencies
    Assumes periodic signals
 Periodic signals (definition):
    Repetitive
    Repetitive
    Repetition occurs at uniformly spaced intervals of time
 Periodic signal is assumed to persist from infinite
  past to infinite future
Fourier Series Representation
 If a signal is periodic,
  the signal can be
  expressed as the sum
  of sine and cosine
  waves of different
  amplitudes and
  frequencies
 This is known as the
  Fourier Series
  Representation of a
  signal
Fourier Series Representation
 Pragmatic Details
    Lowest Fundamental Frequency is 1/T
       T=period sampled by the N samples
    Resolution is 1/T
 Phase and Power
    There exist a phase component and an amplitude
     component to the Fourier series representation
       Using both, it is possible to completely reconstruct the
        waveform.
 Psychophysiologist usually only interested in amplitude
  component
                                 Sine
                                 wave




                                 9.75 Hz
                                 wave
                                 with noise
                                 added


                                 Noise
                                 Only
                                 < 20 Hz



                               Mixture of
Time Domain   Frequency Domain 3 waveforms
  Averaging
  Multiple Epochs
  improves ability
  to resolve signal




Note noise is twice
amplitude of the signal
Lingering details
 In absence of phase information, it is impossible
  to reconstruct the original signal
  Infinite number of signals that could produce the
   same amplitude or power spectrum
 Spectra most often derived via a Fast Fourier
  transform (FFT); a fourier transform of a
  discretely sampled band-limited signal with a
  power of 2 samples
 Windowing: the Hamming or Hanning Taper
Hanning Window – Overlapping epochs by 75% will reduce
effects of smearing
Pragmatic concerns with sample rate

Sample fast enough so no frequencies
 exceed Nyquist
Sample a long enough epoch so that
 lowest frequency will go through at least
 one period
Sample a periodic signal
  if subject is engaging in task, make sure that
   subject is engaged during entire epoch
Lateralization of cognitive specialization

 Damage to left hemisphere interferes with language
  processing & math/analytic tasks
 Damage to right hemisphere interferes with spatial
  relations & gestalt, synthetic tasks
 Galin & Ornstein (1972)
    Ps perform 4 tasks (verbal, spatial vs. mental, motor)
        Verbal = write letter & mentally compose a letter
        Spatial = 2-dimensional geometric pattern to memorize & construct
         pattern with colored blocks; sectioned figure & must id what it is
    EEG from T3/4 and P3/4
    Analyzed power in 1-35 Hz
    Greater right than left activity (inverse) occurred in spatial
    Greater left than right activity occurred in verbal
Concerns with Galin & Ornstein (1972)

 Use of power between 1-35 Hz
 Were verbal & spatial tasks matched on
  difficulty?
 Davidson, Chapman, Chapman, & Henriques
  (1990) used verbal & spatial tasks matched on
  difficulty (mean, SD, alpha) – dot localization &
  word finding
   Found significant task x hemisphere effects for delta,
    theta, alpha, and beta
       Strongest effects for alpha in central & parietal regions
   Less power in most activated hemisphere
   No functional significance of different bands
Frontal lateralization of emotion

Lesion findings
  LH damage -- depression
  RH damage -- mania
EEG studies
   Trait (40+ studies)
   State (25 + studies)
 Left Hypofrontality in Depression




Henriques & Davidson (1991); see also, Allen et al. (1993), Gotlib et al. (1998);
Henriques & Davidson (1990); Reid Duke and Allen (1998); Shaffer et al (1983)
Individual
Subjects’ Data
From Tomarken et al. (1992) – bars refer to alpha power; Ss with more left frontal
activity (inverse of alpha) had greater PA and a trend to lesser NA; groups were
top of bottom quartiles on asymmetry over 2 sessions.
From Tomarken & Davidson (1994) -- Mean standardized Marlowe-Crowne Social
Desirability Scale, State-Trait Anxiety Inventory (STAI) Trait, and Beck Depression
(BDI) scores for the left midfrontal ( n = 12) and right midfrontal ( n = 13) groups.
Groups were top of bottom quartiles on asymmetry over 2 sessions.
The Behavioral Approach System
 sensitive to signals of
   conditioned reward
   nonpunishment
   escape from punishment
 Results in:
   driven pursuit of appetitive stimuli
   appetitive or incentive motivation
   Decreased propensity for depression (Depue &
     Iacono, 1989; Fowles 1988)
Methods for BAS/BIS Study

 Carver & White (1994) BAS/BIS scales
  BIS
      I worry about making mistakes.
      I have very few fears compared to my friends.
  BAS
      When good things happen to me, it affects me strongly.
      I go out of my way to get things I want.
      I crave excitement and new sensations.
 Resting EEG (4 min)
              a     a




                                      b      c




Bars with different superscripts are different at p < .05.
Results of BAS/BIS Study
Harmon-Jones & Allen (1997)


BAS related to relative left frontal activity
   r = .38 , p < .05
   Even when total frontal alpha power was
    entered first in a regression analysis: R2 = .11
    , p < .05, beta = .35
BIS did not relate to frontal asymmetry

 BAS results replicated in Sutton & Davidson
  (1997)
State Changes

 Infants
     Stanger/Mother paradigm (Fox & Davidson,
     1986)
     Sucrose Vs water (Fox & Davidson, 1988)
     Films of facial expressions (Jones & Fox, 1992;
     Davidson & Fox, 1982)

Fox & Davidson (1988)
                  Mean EEG Power for
                   the 3—12 Hz
                   Frequency Band for
                   the Scalp Leads F3
                   and F4 (Referenced
                   to Cz) in 10-Month-
                   Old Infants Who
                   Displayed Smiles
                   With or Without
                   Orbicularis Oculi
Field et al. (1995)
                       Frontal EEG
                        asymmetry scores for
                        infants of depressed
                        and nondepressed
                        women and frontal
                        EEG asymmetry
                        scores for depressed
                        and nondepressed
                        mothers
Models of Asymmetrical Frontal
Activity
Motivational Model
  Left frontal – approach motivation
  Right frontal – withdrawal motivation


Valence Model
  Left frontal – positive affect
  Right frontal – negative affect
Predictions for the two models…

Converge for most emotions
  Positive emotions (interest, joy) are associated
   with approach motivation
  Negative emotions (fear, disgust) are
   associated with withdrawal motivation
But what about ANGER?
Anger as Negative

Often occurs in response to negatively
 evaluated situations
Often evaluated as a negative feeling state
Anger as Approach

Evokes approach action tendencies
  Darwin (1965), Plutchik (1980), Izard (1991)


Anger may underlie offensive aggression
  Blanchard & Blanchard (1984), Lagerspetz
   (1969), Moyer (1976)
Offensive Aggression
       ―angry‖




Defensive Aggression
      ―fearful‖
Anger as Approach
 Trait anger relates to trait behavioral approach
  sensitivity
 at simple correlation level & in simultaneous
  regression controlling for negative affect and
  behavioral inhibition
 BAS example item
   ―I crave excitement and new sensations.‖
   ―It would excite me to win a contest.‖
 Trait anger example items
   ―When frustrated, I let my irritation show‖
   ―Some of my friends think I am a hothead.‖
Anger as Approach

Predicting trait anger
Predictor     partial r             p
BAS           .26                   .001
BIS           .08                   .35
NA            .35                   .001

 from Harmon-Jones (in press, Study 1)
Anger as Approach

Predicting physical aggression
Predictor    partial r p
BAS          .46       .003
BIS          -.53      .001
NA           .60       .001

 from Harmon-Jones (in press, Study 2)
Competing Predictions

If frontal asymmetry reflects valence of
 emotion, then anger should relate to
 increased right frontal activity, because
 anger is a negative emotion
If frontal asymmetry reflects motivational
 direction of emotion, then anger should
 relate to increased left frontal activity,
 because anger evokes approach
 tendencies
Relationship of trait anger and resting
frontal asymmetry
 Assessed relationship between resting frontal
  EEG over 8 min & trait anger
   Buss & Perry (1992) trait questionnaire
   ―When frustrated, I let my irritation show.‖
 Trait anger related to increased left frontal
  activity and decreased right frontal activity
 In 2 samples
    Harmon-Jones & Allen (1998, JPSP)
       Middle school children
    Harmon-Jones (2002, under review)
       College students
 Anger and Asymmetry Correlations




Areas in deep red reflect significant positive
correlations. Areas in blue reflect n.s. negative
correlations.
Controlling for positive affect and negative
 affect, as assessed by PANAS, did not
 alter magnitude of anger & asymmetry
 correlation.
Addressing an Alternative Explanation for Trait
Anger Research

Alternative explanation – persons high in
 trait anger regard anger as a positive
 feeling
   Created Attitudes toward Anger individual differences
    questionnaire; ―I like how it feels when I am mad.‖
Trait anger related to relative left frontal
 activity but attitude toward anger (ATA) did
 not.
ATA did not mediate the anger-asymmetry
 relationship.
                                           Harmon-Jones (under review)
State Anger and
Frontal Asymmetry
Would situationally-induced anger relate to
 relative left frontal activity?
Method
Harmon-Jones & Sigelman (2001, JPSP)

 Told they would participant in two perception
  tasks – person perception & taste perception
 Person perception task – insult manipulation
   participant writes essay on important social issue;
    another ostensible participant gives written feedback on
    essay
   Feedback is neutral or insulting
Neutral Feedback
 Boring                      Thought-Provoking
     1 2 3 4 5 6 7 8 9

 Other P gave moderately positive ratings on
  intelligent, thought-provoking, friendly, logical,
  respectable, & rational.

 Under additional comments:
   ―I can understand why a person would think like
    this.‖
Anger-Inducing Feedback
 Boring                    Thought-Provoking
    1 2       3 4 5 6 7 8 9
 Other P gave negative ratings on intelligent,
  thought-provoking, friendly, logical,
  respectable, & rational.
 Under additional comments:
  ―I can’t believe an educated person would think like
    this. I hope this person learns something while at
    UW.‖
Record EEG immediately after feedback
Taste perception task – aggression
 measure
  participant selects beverage for other
   participant, ―so that experimenter can remain
   blind to type of beverage.‖
  6 beverages; range from pleasant-tasting
   (sweetened water) to unpleasant-tasting (water
   with hot sauce)
Harmon-Jones & Sigelman (2001, JPSP)

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:29
posted:8/18/2011
language:English
pages:69